Please use R$-$Squared as evaluation metric.

$1$ feature selection method.

$2$ machine learning algorithms $($different types, no models$).$

Analysis and results

Preliminary conclusions and ideas to further improve your model for the contest.

$-$ Additional feature selection methods $($even those not reviewed in class$).$

$-$ Other feature engineering techniques

$-$ Cross$-$validation

$-$ Additional machine learning algorithms $($even those not reviewed in class$).$

$-$ Hyper$-$parameter tuning

$-$ Learning curvesDataset Description: The dataset is composed of several next$-$day forecast variables, maximum and

minimum temperatures of present$-$day, and geographic auxiliary variables collected for a period of $5$

years by the Korean Meteorological Service over Seoul, South Korea. The output variable is the next$-$day

average temperature $($NextDayAvTemp$).$

Dataset Variables:

Station: Used weather station number.

Present Tmax: Maximum air temperature between $0$ and $21h$ on the present day $(C).$

Present Tmin: Minimum air temperature between $0$ and $21h$ on the present day $(C).$

NextDayPred RHmin: Forecast of next$-$day minimum relative humidity $(\%).$

NextDayPred$\_$RHmax: Forecast of next$-$day maximum relative humidity $(\%)$

NextDayPred$\_$Tmax lapse: Forecast of next$-$day maximum air temperature applied lapse rate $(C)$

NextDayPred$\_$Imin lapse: Forecast of next$-$day minimum air temperature applied lapse rate $(C)$

NextDayPred$\_$WS: Forecast of next$-$day average wind speed $(\frac{m}{s})$

NextDayPred$\_$LH: Forecast of next$-$day average latent heat flux $($W$/$m$2)$

NextDayPred$\_$CC$1$: Forecast of next$-$day $1$ st $6-$hour split average cloud cover $(0-5h.$

NextDayPred$\_$CC$2$: Forecast of next$-$day $2$nd $6-$hour split average cloud cover $(6-11$ h$)(\%).$

NextDayPred$\_$CC$3$: Forecast of next$-$day $3$rd $6-$hour split average cloud cover $(12-17$ h$)(\%).$

NextDayPred$\_$CC$4$: Forecast of next$-$day $4$th $6-$hour split average cloud cover $(18-23$ h$)(\%).$

NextDayPred$\_$PPT$1$: Forecast of next$-$day $1$ st $6-$hour split average precipitation $(0-5$ h$)(\%).$

NextDayPred