With the D3.js foundation in place, it was time to populate the knowledge graph with real data. This meant diving deep into my academic records and transforming a raw list of courses into a meaningful visualization that accurately represents my Computer Engineering journey at Pontificia Universidad Católica de Chile.
Extracting Data from SIDING
The first challenge was gathering comprehensive course data from UC's academic platform, SIDING (Sistema de Información para la Docencia de Ingeniería). This platform serves as the central hub for all academic information, including the Curricular Tracker - a tool that maps every student's progress through their degree requirements.
The SIDING Experience
SIDING's Curricular Tracker provided me with:
- Complete course history: Every course taken from 2019 to 2024
- Course codes and official names: Essential for accurate representation
- Credit information: Indicating the relative importance of each course
- Prerequisite relationships: Understanding the academic flow
- Grade records: Personal performance context
However, the raw data was just a long list of completed courses with codes like "IIC2133" and "MAT1610" - not immediately meaningful for visualization purposes.
Academic Categorization Process
The next step involved organizing 80+ courses into logical macro-topics that would make sense both academically and professionally. This required understanding how different courses contribute to my overall engineering education.
Defining Macro-Topics
I identified seven key areas that encompass the Computer Engineering curriculum:
1. Basic Engineering (17 courses)
The foundational mathematics, physics, and general engineering courses that every engineer needs:
- Mathematics sequence: Calculus I-III, Linear Algebra, Differential Equations
- Physics fundamentals: Dynamics, Thermodynamics, Electricity & Magnetism
- General engineering: Chemistry, Biology, Economics, Ethics
2. Introductory Computer Engineering (4 courses)
The gateway courses that introduce students to computational thinking:
- Introduction to Programming (IIC1103)
- Discrete Mathematics (IIC1253)
- Advanced Programming (IIC2233)
- Science and Technology in Digital Age (IIC1005)
3. Core Computer Engineering (5 courses)
Essential computer science fundamentals that every software engineer must master:
- Data Structures and Algorithms (IIC2133)
- Computer Architecture (IIC2343)
- Operating Systems and Networks (IIC2333)
- Databases/SQL (IIC2413)
- Distributed Systems (IIC2523)
4. Theoretical Computer Science (5 courses)
The mathematical and theoretical foundations of computing:
- Algorithm Analysis and Design (IIC2283)
- Theory of Automata and Formal Languages (IIC2223)
- Logics for Computer Science (IIC2213)
- Cryptography and Computer Security (IIC3243)
- High Performance Computing (IIC3533)
5. Data Science and AI (8 courses)
My specialization area, focusing on machine learning and data analysis:
- Artificial Intelligence (IIC2613)
- Data Mining (IIC2433)
- Process Mining (IIC3757)
- Recommender Systems (IIC3633)
- Statistical Inference (EYP2114)
- Information Visualization (IIC2026)
- Data Science for Health (IIC3800)
- Advanced Topics in AI (IIC3692)
6. Software Engineering and Development (5 courses)
Practical software development and engineering practices:
- Software Engineering (IIC2143)
- Web Applications (IIC2513)
- Software Architecture (IIC2173)
- Software Development (IIC3143)
- Specialty Project/Capstone (IIC2154)
7. Information Technology and Management (2 courses)
Business and management aspects of technology:
- Information Systems (IIC2713)
- IT Project Management (IIC3113)
Professional Relevance Weighting
The most critical step was assigning weights to each course based on their relevance to my career goals as a Data Scientist and Full-Stack Developer. This involved honest self-reflection about which courses truly impact my daily work and future aspirations.
Weighting Criteria
I developed a relevance scale considering:
- Direct applicability to current projects and job requirements
- Frequency of use in real-world software development
- Foundation strength for advanced topics
- Industry demand and market relevance
- Personal interest and specialization goals
Weight Categories
The courses were assigned relative sizes that translate to visual prominence in the graph:
Extra Large (xl): Core competencies I use daily
- Data Structures and Algorithms
- Databases/SQL
- Web Applications
- Software Engineering
Large (l): Essential skills for my career path
- Artificial Intelligence
- Data Mining
- Advanced Programming
- Software Architecture
Medium-Large (ml): Important foundational knowledge
- Computer Architecture
- Operating Systems
- Distributed Systems
Medium (m): Solid background knowledge
- Discrete Mathematics
- Process Mining
- Software Development
Small (s-xs): Specialized or less directly applicable
- Basic engineering courses
- Theoretical computer science topics
- General education requirements
Implementation Challenges
Balancing Academic vs. Professional Perspective
One of the trickiest aspects was balancing academic importance with professional relevance. For example:
- Calculus III is academically fundamental but rarely used in web development
- Process Mining is highly specialized but directly relevant to my research interests
- Ethics is crucial for professional development but doesn't involve technical skills
Avoiding Bias
I had to be careful not to undervalue courses simply because I found them challenging or less interesting at the time. Some courses that seemed theoretical during study (like Cryptography) proved valuable in understanding modern security practices.
Maintaining Academic Integrity
The visualization needed to accurately represent the comprehensive nature of engineering education, not just highlight my preferred subjects. This meant ensuring adequate representation of all seven macro-topics.
Technical Implementation
The weighted course data was structured as a hierarchical graph:
interface Node {
id: number;
label: string;
size?: "xl" | "l" | "ml" | "m" | "ms" | "s" | "xs";
}This allows the D3.js visualization to render nodes with different sizes based on professional relevance:
.attr("r", (d: any) => {
if (d.size === "xl") return 35;
else if (d.size === "l") return 25;
else if (d.size === "ml") return 20;
// ... and so on
})Results and Insights
The final knowledge graph now contains:
- 82 total courses across 5 years of study
- 7 macro-topics representing different engineering domains
- Weighted visualization highlighting career-relevant skills
- Hierarchical structure showing academic progression
Unexpected Discoveries
The categorization process revealed interesting patterns:
- Data Science courses represent only 10% of total coursework but 40% of career relevance
- Basic Engineering courses, while foundational, have limited direct application
- Software Engineering practices bridge theoretical knowledge and practical skills
- Interdisciplinary courses often prove more valuable than initially expected
Future Enhancements
This comprehensive course mapping enables several planned features:
- Interactive course details with descriptions and projects
- Skill clustering to show related competencies
- Timeline visualization showing academic progression
- Career path mapping connecting courses to job requirements
The transformation from raw SIDING data to weighted knowledge graph demonstrates how academic records can become powerful tools for professional self-reflection and career communication.
Next up: Implementing interactive features and detailed course information to make the knowledge graph truly explorable.