Enabling non-technical users to query and purchase data
The increasing availability of data has offered companies and organizations the opportunity to optimize their business strategies and operations by taking advantage of data-driven decision-making. In this data-intense context, we have seen an increase in the number of non-technical users that need to query, explore, and make sense of large volumes of data. However, for these users, interacting with data becomes a tedious and time-consuming process. Furthermore, in cases where data is not readily available, users frequently make use of data marketplaces to acquire high-quality data. Nevertheless, current marketplaces offer almost no help to non-technical users.
In this thesis, we introduce a novel framework to enable users with limited technical expertise to conveniently (1) communicate their preferences and information needs, (2) formulate database queries without requiring them to be familiar with any query language, (3) evaluate data offers available in a data marketplace, and (4) manage the costs involved in issuing such queries to a marketplace in cases where the available budget is insufficient to embrace the complete target data set. The results obtained by the analyses reported in this thesis demonstrate that the proposed framework can produce near-optimum solutions and provide well-suited support to non-technical users in the tasks of query formulation and data purchase.
database; query formulation; query reverse engineering; data purchase; data marketplace; machine learning; non-technical users