Package de.uni_trier.wi2.procake.similarity.base

Interface SMInterval

  • All Superinterfaces:
    SimilarityMeasure
    All Known Implementing Classes:
    SMIntervalImpl
    public interface SMInterval
extends SimilarityMeasure
    The attribute value in the query and in the case are intervals. We name the query interval $x_i = [x_{ilb},x_{iub}]$, the case interval $y_i = [y_{ilb},y_{iub}]$, and a possible intersection $z_i = [z_{ilb},z_{iub}]$. We have to distinguish the four combinations of the knowledge types.

    The interval in the case means that a valid value exists in this range, but it is not sure where it is. It can't be assumed that valid values are in the intersection of query and case interval. Therefore, the three strategies must be distinguished:

    The optimistic one assumes that valid points are in the intersection. With the imprecise query one point is requested from all the possible points of the precise case. Therefore, we just have to distinguish the following two situations:

    • $z_i \not= \emptyset$: The local similarity is one, because an arbitrary element of $z_i$ or $z_i$ itself can be returned.
    • $z_i = \emptyset$: The local similarity must consider the ``distance'' from $x_{ilb}$ to $y_{iub}$ and $x_{iub}$ to $y_{ilb}$, so that \[sim^*_{Ai}(x_i,y_i) = \max\left(sim_{Ai}(x_{ilb},y_{iub}), sim_{Ai}(x_{iub},y_{ilb})\right)\]

    Summarizing, a local similarity measure can be defined as: \[ sim^*_{Ai}(x_i, y_i) = \left\{ \begin{array}{c@{\quad}l} 1 & \mbox{if} \quad z_i \not= \emptyset\\ \max\left(\begin{array}{l} sim_{Ai}(x_{ilb},y_{iub}),\\ sim_{Ai}(x_{iub},y_{ilb}) \end{array}\right) & \mbox{otherwise} \end{array} \right. \]

    The pessimistic strategy assumes that no valid point is in the intersection which leads to the local similarity \[sim^*_{Ai}(x_i, y_i)=\\underset{\forall x_{is}\in x_i, y_{it}\in y_i}{\min}\left\{sim_{Ai}(x_{is},y_{it})\right\}\]

    The average strategy calculates the probability that a valid point is in the intersection $`z_i`$, e.g. by calculating the relation between the intersection size and the case interval size. Consequently, the similarity measure is defined as: $`sim^*_{Ai}(x_i, y_i) = \begin{cases} \frac{|y_{iub}-y_{ilb}|}{|z_{iub}-z_{ilb}|} & \text{, if } z_i =\not \emptyset \\ 0 &, otherwise \end{cases}`$

    Author:
    Rainer Maximini

    • Field Detail

      • DEFAULT_STRATEGY

        static final Strategy DEFAULT_STRATEGY
        Default strategy for similarity computations is .

    • Method Detail

      • getOrderName

        String getOrderName()

      • setSimilarityToUse

        void setSimilarityToUse​(String newValue)
        In general, the element objects of the interval are compared with their default similarity measure. But in some situations it can be necessary to use another similarity measure for the elements of an interval. Therefore, it exists the possibility to specify a similarity measure name that should be used instead. For each DataObject a similarity measure with that name should exist. Otherwise, the comparision of objects are ignored.

        Summarizing:

        • If the newValue is null the default measures of the objects are used. This is the default behaviour.
        • If the newValue is the name of a similarity measure, for each data class whose objects can be occured in the interval a similarity measure with this name must exist.
        Parameters:
        newValue - The name of the similarity measure that should be used for the elements.

      • setStrategy

        void setStrategy​(Strategy strategy)

      • removeOrder

        void removeOrder()
        Similarity measure for intervals can work also on the natural order of a AtomicDataClass if available. Therefore, it is possible to remove an already assigned order.