Handling qualitative preferences in SPARQL over virtual ontology-based data access

With the increase of data volume in heterogeneous datasets that are being published following Open Data initiatives, new operators are necessary to help users to find the subset of data that best satisfies their preference criteria. Quantitative approaches such as top-k queries may not be the most appropriate approaches as they require the user to assign weights that may not be known beforehand to a scoring function. Unlike the quantitative approach, under the qualitative approach, which includes the well-known skyline, preference criteria are more intuitive in certain cases and can be expressed more naturally. In this paper, we address the problem of evaluating SPARQL qualitative preference queries over an Ontology-Based Data Access (OBDA) approach, which provides uniform access over multiple and heterogeneous data sources. Our main contribution is Morph-Skyline++, a framework for processing SPARQL qualitative preferences by directly querying relational databases. Our framework implements a technique that translates SPARQL qualitative preference queries directly into queries that can be evaluated by a relational database management system. We evaluate our approach over different scenarios, reporting the effects of data distribution, data size, and query complexity on the performance of our proposed technique in comparison with state-of-the-art techniques. Obtained results suggest that the execution time can be reduced by up to two orders of magnitude in comparison to current techniques scaling up to larger datasets while identifying precisely the result set.

Learning reward functions from diverse sources of human feedback: Optimally integrating demonstrations and preferences

Reward functions are a common way to specify the objective of a robot. As designing reward functions can be extremely challenging, a more promising approach is to directly learn reward functions from human teachers. Importantly, data from human teachers can be collected either passively or actively in a variety of forms: passive data sources include demonstrations (e.g., kinesthetic guidance), whereas preferences (e.g., comparative rankings) are actively elicited. Prior research has independently applied reward learning to these different data sources. However, there exist many domains where multiple sources are complementary and expressive. Motivated by this general problem, we present a framework to integrate multiple sources of information, which are either passively or actively collected from human users. In particular, we present an algorithm that first utilizes user demonstrations to initialize a belief about the reward function, and then actively probes the user with preference queries to zero-in on their true reward. This algorithm not only enables us combine multiple data sources, but it also informs the robot when it should leverage each type of information. Further, our approach accounts for the human’s ability to provide data: yielding user-friendly preference queries which are also theoretically optimal. Our extensive simulated experiments and user studies on a Fetch mobile manipulator demonstrate the superiority and the usability of our integrated framework.

Preference queries over taxonomic domains

When composing multiple preferences characterizing the most suitable results for a user, several issues may arise. Indeed, preferences can be partially contradictory, suffer from a mismatch with the level of detail of the actual data, and even lack natural properties such as transitivity. In this paper we formally investigate the problem of retrieving the best results complying with multiple preferences expressed in a logic-based language. Data are stored in relational tables with taxonomic domains, which allow the specification of preferences also over values that are more generic than those in the database. In this framework, we introduce two operators that rewrite preferences for enforcing the important properties of transitivity, which guarantees soundness of the result, and specificity, which solves all conflicts among preferences. Although, as we show, these two properties cannot be fully achieved together, we use our operators to identify the only two alternatives that ensure transitivity and minimize the residual conflicts. Building on this finding, we devise a technique, based on an original heuristics, for selecting the best results according to the two possible alternatives. We finally show, with a number of experiments over both synthetic and real-world datasets, the effectiveness and practical feasibility of the overall approach.

Incremental Elicitation of Rank-Dependent Aggregation Functions based on Bayesian Linear Regression

We introduce a new model-based incremental choice procedure for multicriteria decision support, that interleaves the analysis of the set of alternatives and the elicitation of weighting coefficients that specify the role of criteria in rank-dependent models such as ordered weighted averages (OWA) and Choquet integrals. Starting from a prior distribution on the set of weighting parameters, we propose an adaptive elicitation approach based on the minimization of the expected regret to iteratively generate preference queries. The answers of the Decision Maker are used to revise the current distribution until a solution can be recommended with sufficient confidence. We present numerical tests showing the interest of the proposed approach.

A Cognitive Information Retrieval Using POP Inference Engine Approaches

The focus of this chapter is to design a cognitive information retrieval (CIR) framework using inference engine (IE). IE permits one to analyze the central concepts of information retrieval: information, information needs, and relevance. The aim is to propose an inference engine in which adequate user preferences are considered. As the cognitive inference engine (CIE) approach is involved, the complex inquiries are required to return more important outcomes as opposed to customary database questions which get irrelevant and unsolicited responses or results. The chapter highlights the framework of a cognitive rule-based engine in which preference queries are dealt with while keeping in mind the intention of the user, their performance, and optimization.