Narrow Results

This paper presents the concept of kernels to address the complexity of solving big-data applications. Their solution strategies often require evaluating domain-dependent subspaces on the big data and selecting the best result. As the data space in these problems is so vast that it is infeasible to scan the data once, we need domain-specific methods for identifying promising and manageable subspaces with a high density of solutions before looking for individual ones. To this end, we introduce kernels to represent some properties of the statistical quality, average density, or probability of solutions in a subspace to prune subspaces with suboptimal kernels. We classify various past approaches based on their analysis methods and illustrate each by an example. Our results confirm that kernels can effectively harness the complexity of solving big-data applications.

For a company, it is important to know which products to launch to the market that may get the maximal profit. To achieve this goal, companies not only need to consider these products’ features, but also need to analyze how customers make their purchase decisions. For most customers, the price of a product is the most important purchase factor. If the price of a product can be adjusted, the purchase decision of a customer may change. With different price settings, we can speculate on the expected number of customers and the profit of the products. Motivated by this, we want to find the most profitable products among the candidates for the company. A distance-based adoption model can be used to evaluate the expected customers for products at different prices. The computational cost is high in two parts. One is the computational cost of obtaining the most profitable information on each set of candidate products. Another part is that many candidate product combinations need to be calculated. To tackle the computation problem, we propose two strategies. One is to avoid considering all possible price settings. The other is to avoid processing all possible subsets of the candidate products. Experimental results reveal the efficiency of our strategies.