Query Answer Reformulation over Knowledge Bases

The answer of a query, submitted to a database or a knowledge base, is often long and may contain redundant data. The user is frequently forced to browse through a long answer or refine and repeat the query until the answer reaches a manageable size. Without proper treatment, consuming the answer may indeed become a tedious task. This article then proposes a process that modifies the presentation of a query answer to improve the quality of the user’s experience in the context of an RDF knowledge base. The process reorganizes the original query answer by applying heuristics to summarize the results and to select template questions that create a user dialog that guides the presentation of the results. The article also includes experiments based on RDF versions of MusicBrainz, enriched with DBpedia data, and IMDb, each with over 200 million RDF triples. The experiments use sample queries from well-known benchmarks.

The Shapley Value of Tuples in Query Answering

We investigate the application of the Shapley value to quantifying the contribution of a tuple to a query answer. The Shapley value is a widely known numerical measure in cooperative game theory and in many applications of game theory for assessing the contribution of a player to a coalition game. It has been established already in the 1950s, and is theoretically justified by being the very single wealth-distribution measure that satisfies some natural axioms. While this value has been investigated in several areas, it received little attention in data management. We study this measure in the context of conjunctive and aggregate queries by defining corresponding coalition games. We provide algorithmic and complexity-theoretic results on the computation of Shapley-based contributions to query answers; and for the hard cases we present approximation algorithms.

Technical Perspective for

When a data analyst runs some query to analyze her data, she often wants to ask some follow-up questions, about the result of the query. Why-questions take many shapes, and occur in many scenarios. Why is a particular tuple in the answer? Why is it not in the answer? Why is this graph decreasing? Why did we observe a sudden burst of error messages in online monitoring? Database researchers have noted the need for why-questions, and the literature contains several approaches, mostly tailored to specific applications. Despite the interest and the work in this area, there is currently no consensus of what an explanation to a query answer should be, and how one should compute it.

Improving the Quality of the User Experience by Query Answer Modification

This paper proposes a process that modifies the presentation of a query answer to improve the quality of the user’s experience. The process is particularly useful when the answer is long and repetitive. The process reorganizes the original query answer by applying heuristics to summarize the results and to select template questions that create a user dialog that guides the presentation of the results.

Constructing Datasets for Multi-hop Reading Comprehension Across Documents

Most Reading Comprehension methods limit themselves to queries which can be answered using a single sentence, paragraph, or document. Enabling models to combine disjoint pieces of textual evidence would extend the scope of machine comprehension methods, but currently no resources exist to train and test this capability. We propose a novel task to encourage the development of models for text understanding across multiple documents and to investigate the limits of existing methods. In our task, a model learns to seek and combine evidence — effectively performing multihop, alias multi-step, inference. We devise a methodology to produce datasets for this task, given a collection of query-answer pairs and thematically linked documents. Two datasets from different domains are induced, and we identify potential pitfalls and devise circumvention strategies. We evaluate two previously proposed competitive models and find that one can integrate information across documents. However, both models struggle to select relevant information; and providing documents guaranteed to be relevant greatly improves their performance. While the models outperform several strong baselines, their best accuracy reaches 54.5% on an annotated test set, compared to human performance at 85.0%, leaving ample room for improvement.

A Safe-Region Approach to a Moving k-RNN Queries in a Directed Road Network

In road networks, [Formula: see text]-range nearest neighbor ([Formula: see text]-RNN) queries locate the [Formula: see text]-closest neighbors for every point on the road segments, within a given query region defined by the user, based on the network distance. This is an important task because the user's location information may be inaccurate; furthermore, users may be unwilling to reveal their exact location for privacy reasons. Therefore, under this type of specific situation, the server returns candidate objects for every point on the road segments and the client evaluates and chooses exact [Formula: see text] nearest objects from the candidate objects. Evaluating the query results at each timestamp to keep the freshness of the query answer, while the query object is moving, will create significant computation burden for the client. We therefore propose an efficient approach called a safe-region-based approach (SRA) for computing a safe segment region and the safe exit points of a moving nearest neighbor (NN) query in a road network. SRA avoids evaluation of candidate answers returned by the location-based server since it will have high computation cost in the query side. Additionally, we applied SRA for a directed road network, where each road network has a particular orientation and the network distances are not symmetric. Our experimental results demonstrate that SRA significantly outperforms a conventional solution in terms of both computational and communication costs.