In this research, we studies and analyses the data from the Household Socio-Economic Survey 2009 of the Office of National Statistics, Thailand. We use K-means algorithm to cluster the expenditure of the population. By using DB Index and SD Validity, Index, we found that the appropriate number of clusters is three clusters. Then we use association rule technique to determine the relationship between variables. The results show that the association rules are similar among clusters. For example, the average monthly household income is related to the average monthly household cost. Besides, household size is related to number of earners per household. Furthermore, the monthly tobacco cost for a household is associated with a number of members who are entitled to reimbursement for medical expenses.