An Approximation Scheme for Data Monetization

The unprecedented rate at which data is being generated has led to the growth of markets where valuable datasets are bought and sold. A salient feature this market that a data-buyer (agent) endowed with multi-dimensional private information, namely her ``ideal'' record she values most how valuation for given changes as its distance from ideal changes. Consequently, revenue-maximization problem faced by data-seller (principal), who serves multiple buyers, mechanism-design problem, well-recognized difficult solve. Our main result in paper an approximation scheme guarantees revenue within close positive amount optimal desired. generates posted-price menu consisting set item-price pairs -- each entry consists item, i.e., records dataset, price corresponding item. As tradeoff, length resulting increases desired guarantee gets closer zero. For convenience practice, data-sellers may want ability limit used scheme. To facilitate this, we extend our analysis obtain general any length. We also demonstrate seller can exploit buyers' preferences generate intuitive useful rules thumb effective practical implementation