Github

點我去看程式碼

1. System Analysis

• Objective: Create an online NBA betting system named Oin.

• Users:

  • User: Regular bettors who can deposit money, place bets, update/cancel bets, and check results.
  • Admin: Operators who update odds, settle games, and manage user bet records.

2. System Functions

User Functions

• View season rankings (current + past 10 seasons).
• View personal account info and bet history.
• Check match details, betting options, and odds.
• Search games by date.
• Deposit money and place bets.
• Update personal info and cancel bets before matches start.

Admin Functions

• Query top 5 ROI users.
• View all bets placed on a specific day.
• Open/close betting options by setting odds.
• Settle games and update balances automatically.

System Commands

• Update odds.
• Update game scores.
• Add new games.

3. System Design

• ER Diagram: 8 entities + 1 weak entity (e.g., Gambler, Game, Bet Odds, Stadium, Season, Team).
• Relational Schema: 10 tables (GAMBLER, CASHFLOW_RECORD, GAMBLER_BETS, BET_ODDS_RECORD, etc.).
• Normalization: Schema satisfies 1NF → 4NF (no partial, transitive, or multi-valued dependencies).

4. System Implementation

• Data Sources: NBA data from Kaggle (games, seasons, teams, stadiums); synthetic data for bets, odds, and cashflows.

• SQL Functions:

  • Users: place/cancel bets, query games, view odds, check standings.
  • Admin: open bets, settle results, query top users.
  • System: update odds, scores, and add matches.

• Optimization: Indexing home_team_id and away_team_id improved query efficiency for past match history.

• Transaction Management: Each insert/update wrapped in transaction with commit/rollback.

• Concurrency Control: Not required since bets are independent and only open games are available.

5. Reflections

• Building a database system gave strong sense of achievement.
• Required careful planning to avoid cascading errors.
• Learned practical skills beneficial for future system development.