Shapes are a concise way to describe temporal variable behaviors. Some com- monly used shapes are spikes, sinks, rises, and drops. A spike describes a set of variable values that rapidly increase, then immediately rapidly decrease. The variable may be the value of a stock or a person's blood sugar levels. Shapes abstractly describe a variable's behavior. Details such as the height of a spike or its rate in- crease, are lost in the abstraction. These hidden details make it difficult to define shapes and compare one instance to another. For example, what attributes can be used to define a spike's behavior? And what attributes of a spike determine its "spikiness"? The ability to define and compare shapes is important because it allows shapes to be identified and ranked, according to an attribute of interest. A lot of work has been done in the area of shape identification through pattern match- ing and other data mining techniques, but ideas combining the identification and comparison of shapes have received less attention.
This dissertation fills the gap by presenting a set of shapes and their attributes, by which they can be identified, compared, and ranked. Neither the set of shapes, nor their attributes presented in this dissertation are exhaustive, but it provides an example of how a shape's attributes can be used for identification and compar- ison. Spikes, sinks, rises, drops, lines, plateaus, valleys, and gaps are the shapes presented in this dissertation. Several attributes for each shape are identified and defined. These attributes will be the basis for constructing definitions that identify a particular behavior of a shape and allow it to be ranked.
The second contribution of this work is an information visualization tool, TimeSearcher: Shape Search Edition (SSE), which allows users to explore data sets using the identification and ranking ideas, presented in this dissertation. Case studies were performed to evaluate the benefit of shape identification and ranking in different data sets. Four case studies were performed with a single user, exploring network traffic data and X-ray diffraction data.