CONCLUSION
The data mining procedure based on association rule
mining for extracting relationships among climate parameters
over Cuddalore station was applied to extract the intense
summer day (hot day) patterns during summer months. The
proposed data mining methodology is more useful to apply
with threshold values. As evidenced in the results, the
methodology is suitable for monitoring and predicting the
temperature days 48 hours ahead. This method promises to be
a useful one for tropical coastal stations. By anticipating the
extreme summer temperature, the day to day practice will also
be planned in advance based on human comfort.