**3.1 Visualising inconsistencies in objects with many attribute values pattern**

A dataset contains data about real world objects. These data contains objects which are associated to attributes and the attributes can be associated to single or many values. Real world objects 'G' such as house, book, car, and television are associated with different attributes 'M' which may have many values 'V'. A book (object) for example, can have colour (attribute) which can be black, white or brown (values). It can be established that particular object (g ∈ G) is associated with an attribute (m ∈ M) which contains many values. For example, a name (object) has marital status (attribute) such as married or single (values). Contradictory data can exist in a dataset when there is conflicting information such that an object (g ∈ G) that is associated with an attribute (m ∈ M), contains contradictory values such that m is associated with A and ¬A. An experiment (object) for example, can be associated with outcome (attribute) such as neutral, high, or low (values). A student (object) took a course (attribute) whose values can be absent, pass or fail. Some of the many valued attribute are likely to be mutually exclusive and should conform to mutual exclusion rule. The mutual exclusion rule can simply be stated that real world objects whose attribute values are mutually exclusive (meaning more than one attribute values cannot be associated with the object at once) are contradictory. Also, any attribute which do not contain the expected values is said to contain missing data.

Two open source tools are presented in this chapter to explain how to visualise inconsistencies in objects with many attribute values pattern namely ConTra and Datax. ConTra is discussed in an earlier publication [20] by some of the authors of this chapter and it is also discussed herein. Datax is another tool for highlighting inconsistency in patterns through mining and depicting missing data is presented in Section 3.12.
