Sentence examples for diverse subsets from inspiring English sources

Figure 8a shows that both schemes locate diverse subsets with diversity comparable to the subsets located by Greedy heuristic.

However, these savings in cost are at the expense of decrease in the diversity values of diverse subsets generated using adaptive model built with higher values of (gamma ) as shown in Fig. 9b.

Thus, AdOr-BF tries to find diverse subsets with higher diversity at the expense of higher computational cost.

Specifically, small diverse subsets tend to exhibit higher diversity, whereas the value of diversity decreases with increasing the size of S. Hence, to ascertain the causal effect of diverse subset size k upon the value of the diversity function f(S, d), we assume the value of the diversity function to be the dependent variable, where the diverse subset size is the independent variable.

The regression model is then used to predict future values of diversity function for the selection of diverse subsets across various queries.

The challenge of computing diverse subsets for each query (Q_i) in ({mathbb {M}}) is that those diverse subsets cannot be computed incrementally.

In particular, we employ an adaptive regression model to estimate the diversity of a diverse subset.

In particular, AdOr employs an adaptive model-based diversification method to estimate the diversity of a diverse subset and hence selects diverse results without scanning all the query results.

The central element of our approach is the use of a probabilistic model to estimate the diversity value of a diverse subset for future iterations of Greedy algorithm.

Among the AdOr methods, Adaptive-BF performs better in terms of achieved diversity (Fig. 12b) because in each iteration, Adaptive-BF selects the one result providing the highest diversity if added to the diverse subset.

Thus, with each new addition to the diverse subset the marginal gain in diversity decreases, which is consistent with the study done in previous work [27].