Outliers Detection on Educational Data using Fuzzy Association Rule Mining

Most of the mining techniques have only concerned with interesting patterns. However, in the recent years, there is an increasing demand in mining Unexpected Items or Outliers or Rare Items. Several application domains have realized the direct mapping between outliers in data and real world anomalies that are of great interest to an analyst. Outliers represents semantically correct but infrequent situationin a database. Detecting outliers allows extracting useful and actionable knowledge to the domain experts. In Educational Data, outliers are those students who have secured scores deviated so much from the average scores of other students. The educational data are Quantitative in nature. Any mining technique on quantitative data will partition the quantitative attributes with unnatural boundaries which lead to overestimate or underestimate the boundary values. Fuzzy logic handles this in a more realistic way. Knowing the threshold values apriori is not possible, hence our method uses dynamically calculated Support and Rank measures rather than predefined values. Our method uses a modified Fuzzy Apriori Rare Item sets Mining (FARIM) algorithm to detect the outliers (weak student). This will help the teachers in giving extra coaching for the weak students.