Big_data.docx

BIG DATA-NAÏVE: CLOUDY Yes

WINDY No

HUMID Moderate

DRY Yes

RAINFALL No

Yes

Yes

No

No

Yes

Yes

No

High

Yes

Yes

No

Yes

Moderate

Yes

Yes

No

No

No

No

No

No

Yes

High

Yes

Yes

No

Yes

High

No

No

Yes

Yes

Moderate

Yes

Yes

Determine using suitable classifier? CLOUDY

WINDY

HUMID

DRY

RAINFALL

Yes

No

Moderate

No

?

Sol: P(Rainfall=Y)=5/8 P(Rainfall=N) = 3/8 CLOUDY: P(Cloudy=Y| Rainfall=Y) =3/5 P(Cloudy=Y|Rainfall=N) =1/3 P(Cloudy=N|Rainfall=Y) = 2/5 P(Cloudy=N|Rainfall=N) =2/3 WINDY: P(Windy=Y|Rainfall=Y) =4/5 P(Windy=Y|Rainfall=N) =1/3 P(Windy=N|Rainfall=Y) =1/5 P(Windy=N|Rainfall=N) =2/3

HUMID: P(Humid=Moderate|Rainfall=N) = 1/3 P(Humid=No|Rainfall=N) = 1/3 P(Humid=High|Rainfall=N) = 1/3 P(Humid=Moderate|Rainfall=Y) =2/5 P(Humid=No|Rainfall=Y) =2/5 P(Humid=High|Rainfall=Y) =2/5 DRY: P(Dry=Y|Rainfall=Y) = 4/5 P(Dry=Y|Rainfall=N) =1/3 P(Dry=N|Rainfall=Y) = 1/5 P(Dry=N|Rainfall=N) = 2/3

Given question: P(given conditions|Rainfall=Y) = P(rainfall=Y)*P(Cloudy=Y|rainfall=Y)*P(Windy=No|rainfall=Y)*P(Humid=Moderate|rainfall=Y)*P(Dry=No|rai nfall=Y)=0.006 P(given conditions|Rainfall=N) = P(rainfall=N)*P(Cloudy=Y|rainfall=N)*P(Windy=No|rainfall=N)*P(Humid=Moderate|rainfall=N)*P(Dry=No|r ainfall=N)=9.2x10-3 Therefore, Prediction says no Rainfall occurs for the given condition