Meetings

.gitignore

@@ -5,3 +5,4 @@ venv/
 *.db
 *.idea
 *.ruff_cache
+.DS_Store

.markdownlint.yml

@@ -1,3 +1,4 @@
 ignore:
   - venv
   - .github
+MD033: false

.vscode/settings.json

@@ -122,5 +122,11 @@
       "source.fixAll.ruff": "explicit",
       "source.organizeImports.ruff": "explicit"
     }
-  }
+  },
+  "githubPullRequests.ignoredPullRequestBranches": [
+    "main"
+  ],
+  "cSpell.words": [
+    "Eredivisie"
+  ]
 }

0_domain_study/README.md

@@ -1 +1,125 @@
-# Domain Research
+# ⚽ Background Research: Football as a Domain for Data Science
+
+## 1. Introduction
+
+Football (or soccer) is the most popular sport globally, with an estimated 5
+billion fans and millions of professional and amateur matches played annually.
+Its popularity and data-rich environment make it a prime domain for data science
+research. With the rise of performance analytics, clubs, analysts, and fans
+increasingly rely on data to evaluate player performance, team strategies,
+and match outcomes.
+
+---
+
+## 2. Why Football is a Great Domain for Data Science
+
+### 📈 Data Availability
+
+Football data is widely available from open platforms such as:
+
+- [FBref](https://fbref.com)
+- [Kaggle](https://www.kaggle.com)
+- [Understat](https://understat.com)
+- [Sofascore](https://www.sofascore.com)
+- APIs like [Fotmob](https://www.fotmob.com)
+
+These platforms provide granular match data including:
+
+- Player statistics
+- Match events (goals, fouls, cards)
+- Expected goals (xG)
+- Pass maps
+- Team formations
+
+### 🔍 Rich Analytical Opportunities
+
+Football involves many components that are quantifiable, such as:
+
+- Player performance metrics
+- Tactical formations
+- Physical tracking (via GPS or optical systems)
+- Match outcomes and trends over time
+
+This richness enables application of various analytical techniques:
+
+- Descriptive statistics
+- Predictive modeling
+- Clustering and classification
+- Time-series and spatial analysis
+
+---
+
+## 3. Key Areas of Football Data Science Research
+
+### 3.1 Player Performance Analysis
+
+- Predicting top scorers based on historical metrics
+- Analyzing passing accuracy, dribbles, and defensive contributions
+- Valuing players through regression and machine learning models
+
+### 3.2 Team Strategy and Match Outcome Prediction
+
+- Studying formations and their effectiveness
+- Predicting match results based on pre-match stats
+- Understanding home vs away advantages
+
+### 3.3 Injury and Load Management
+
+- Using GPS tracking data to monitor fatigue
+- Predicting injury risks using machine learning
+
+### 3.4 Fan Engagement and Sentiment Analysis
+
+- Mining social media to analyze public opinion on players or matches
+- Visualizing trends during major tournaments
+
+---
+
+## 4. Examples of Real-World Applications
+
+### 🔬 Clubs & Organizations
+
+- Top clubs (e.g., Liverpool FC, Manchester City) use advanced analytics for
+recruitment and in-match strategy.
+- National teams use data for opponent scouting and tactical planning.
+
+### 📺 Media and Broadcasting
+
+- Broadcasters use data-driven visualizations to enhance storytelling.
+- Fantasy leagues (e.g., Fantasy Premier League) rely heavily on data models.
+
+### 🧪 Academia & Research
+
+- Researchers study fairness, match scheduling, and referee bias.
+- Universities collaborate with clubs for performance modeling and AI applications.
+
+---
+
+## 5. Challenges in Football Data Science
+
+| Challenge                 | Description                                      |
+|--------------------|--------------------------------------------------------- |
+|Data Licensing| Official datasets (e.g., Opta, StatsBomb) can be expensive or restricted.|
+|Tactical Complexity| Tactical nuance is hard to quantify with raw stats alone.|
+|Interpretability| Model predictions must be explainable for coaches and analysts.|
+|Data Imbalance| Many players have few minutes of play, skewing data distribution.|
+
+---
+
+## 6. Conclusion
+
+Football offers a fertile ground for data science exploration. With accessible
+data and a passionate global audience, it's ideal for both academic research
+and portfolio-building projects. From performance metrics to predictive models,
+football allows for the application of a wide range of data science tools in a
+real-world, high-impact context.
+
+As the game evolves, data science will continue to play a central role in shaping
+how we understand and experience football.
+
+**📝 Note:**  
+This folder and its contents will be continuously updated as we expand our
+understanding of football analytics. Expect new papers, tools, and insights to
+be added regularly.
+
+---

0_domain_study/summary_of_group_problem_domain.md

@@ -0,0 +1,119 @@
+# ✅ Milestone 1: Problem Identification  
+
+**Project Title:** *Predicting Post-Transfer Performance of Football
+Players from Lower to Top-Tier Leagues*  
+**Team:** CDSP Group 23 – Hamid, Abdul Qader, Tibyan, Khusro, Saeed  
+
+---
+
+## 🧠 1. Understanding the Problem Domain (with Systems Thinking)
+
+In professional football, **recruiting the right player at the right time** can
+make or break a team’s season. Top-tier clubs such as those in the **Premier
+League, La Liga, or Bundesliga** regularly scout players from **lower-tier leagues**
+ like the **Eredivisie or Championship**, hoping to find breakout stars.
+
+But despite millions spent on transfers, some players underperform. This makes
+ scouting risky, and clubs often ask:
+
+> *"Can we better predict how well a player from a smaller league will adapt and
+> succeed at a higher level?"*
+
+Our project focuses on this critical transition: **from lower-tier leagues to
+ elite-level football**, and **which individual player attributes**
+ (e.g., height, age, speed, experience) are **strongest at predicting success**.
+
+Using **systems thinking**, we view the player’s transfer as a change in the
+ environment. Key components of the system include:
+
+- **Input Factors:** Player’s physical stats, performance history, league background.
+- **Transformation:** New league conditions – higher competition, intensity, tactics.
+- **Outputs:** Post-transfer success (goals, assists, match ratings, minutes played).
+- **Feedback Loops:** Performance influences playing time, fan expectations,
+   market value.
+
+---
+
+### 🌟 Team Insights (Personal Experiences)  
+
+Our diverse perspectives shaped this research question:
+
+<!-- markdownlint-disable MD013 -->
+
+| Member        | Domain Passion                  | Key Observation                          |  
+|---------------|---------------------------------|------------------------------------------|  
+| **Abdul Qader** | Football Analytics             | "High-fee transfers often fail to justify their price tags—what attributes *actually* predict success?" |  
+| **Hamid**      | Football Statistics            | "xG and possession stats dominate analysis, but physical adaption matters too." |  
+| **Khusro**     | Sports Strategy                | "Substitution timing changes games—could player attributes affect adaptation speed?" |  
+| **Tibyan**     | Behavioral Analysis            | "Crowd bias exists—does player physique influence referee decisions in new leagues?" |  
+| **Saeed**      | Sports Medicine & Data         | "Injury risk models exist, but not for *transitioning* players between leagues." |  
+
+<!-- markdownlint-enable MD013 -->
+
+---
+
+### ❓ Research Question  
+
+> **Which individual attributes—such as height, pace, age, or previous league
+> experience—most strongly predict a player's successful performance after
+> moving from a lower-tier league (e.g., Eredivisie) to a top-tier league
+> (e.g., Premier League)?**
+---
+This question is **actionable, measurable, and relevant** to stakeholders in football
+
+---
+
+### 💡 Why This Question?
+
+1. **Solves a Real Problem**  
+2. **Unique Angle**  
+   > This is not just about talent — it’s about **transition readiness**, blending:
+   - 📊 *Performance stats*
+   - 🧬 *Physical traits*
+   - 🎓 *Experience/age*
+
+---
+
+These perspectives helped us select a research question that is **both practical
+ and meaningful** in the current football landscape.
+
+---
+
+## 🧩 5. Stakeholders and Relevance
+<!-- markdownlint-disable MD013 -->
+| Stakeholder              | Why it Matters |
+|--------------------------|----------------|
+| ⚽ **Football Scouts & Clubs**   | Can use results to prioritize scouting criteria and reduce risky transfers. |
+| 📈 **Data Analysts & Journalists** | Insights into player development, talent scouting, and sports analytics. |
+| 🧑‍💻 **Fans & Fantasy Managers**    | Helps anticipate which new signings are likely to succeed. |
+| 🧠 **Players & Agents**           | Can benchmark attributes and expectations when moving between leagues. |
+
+---
+
+## 📝 6. Summary
+
+This project takes a real-world football question—**what makes a transfer
+ succeed?**—and applies a **data-driven, human-centered approach** to answering
+  it. We’re not only learning about modeling, but also building insights that
+   clubs, analysts, and fans might genuinely use.
+
+Our team is excited to explore this question further in the coming milestones!
+
+---
+
+### 🗺️ Systems View
+
+```mermaid
+graph TD
+    A[Lower-Tier Player Profile] --> B{{Transfer Decision}}
+    B --> C[Physical Traits: Height, Pace]
+    B --> D[Technical Metrics: xG, Pass Accuracy]
+    B --> E[Experience Factors: Age, League Tier]
+    C --> F[Top-League Performance]
+    D --> F
+    E --> F
+    F --> G{{Impact on Stakeholders}}
+    G --> H[Scouts: Smarter transfers]
+    G --> I[Agents: Stronger negotiations]
+    G --> J[Fans & Analysts: Realistic expectations]
+  ```