/Classification-of-Football-Teams-on-Relegation-Battle

It is a project using Oracle PL/SQL .

• Task Approach:

There are different football leagues where each season different teams participate representing various clubs. Throughout one season, the poorly performing teams have to face relegation battles and in some leagues, the three teams with least points are relegated while some others survive. In this project work, using the K Nearest Neighbor classification approach, teams will be classified as Relegation Survivor or Relegated based on different criteria in regards to the information from previous seasons.

• Database Schema:

-> Global Schema

Teamlist(teamID, teamName , country, seasonId ) PerformanceList(perID, teamID, win , lost , draw, gf ,ga,points ) RelegationBattle(relID,perID, rbStatus) RecordList(recID,win,lost,draw,gf,ga,points, predictedStatus)

-> Fragmentation Schema

Teamlist1= SLSeasonID=1 Teamlist Teamlist2= SLSeasonID=2 Teamlist Teamlist3= SLSeasonID=3 Teamlist PerformaceList1= SLpoints<35 PerformanceList PerformaceList2= SLpoints>=35 PerformanceList

-> Allocation Schema

Teamlist1 , Teamlist2 , PerformaceList1, PerformaceList2, RelegationBattle, Record at sites 1 , 2

• Significance of Distributed Database System:

While working on classification approach, the most important task is to maintain the data properly so that the approach can make proper predictions. Distributed database systems allow data decentralization which can help with classification tasks. Scaling can be done more conveniently in distributed systems without making huge changes to the whole system. Data maintenance, expansion, reliability can be ensured more efficiently for classification tasks with distributed systems compared to centralized systems