Andreas Thalhammer

Browsing DBpedia Entities with Summaries

The term “Linked Data” describes online-retrievable formal descriptions of entities and their links to each other. Machines and humans alike can retrieve these descriptions and discover information about links to other entities. However, for human users it becomes difficult to browse descriptions of single entities because, in many cases, they are referenced in more than a thousand statements.

LinkSUM: Using Link Analysis to Summarize Entity Data

The amount of structured data published on the Web is constantly growing. A significant part of this data is published in accordance to the Linked Data principles. The explicit graph structure enables machines and humans to retrieve descriptions of entities and discover information about relations to other entities. In many cases, descriptions of single entities include thousands of statements and for human users it becomes difficult to comprehend the data unless a selection of the most relevant facts is provided.

PageRank on Wikipedia: Towards General Importance Scores for Entities

Link analysis methods are used to estimate importance in graph-structured data. In that realm, the PageRank algorithm has been used to analyze directed graphs, in particular the link structure of the Web. Recent developments in information retrieval focus on entities and their relations (i.e., knowledge graph panels). Many entities are documented in the popular knowledge base Wikipedia. The cross-references within Wikipedia exhibit a directed graph structure that is suitable for computing PageRank scores as importance indicators for entities. In this work, we present different PageRank-based analyses on the link graph of Wikipedia and according experiments. We focus on the question whether some links—based on their context/position in the article text—can be deemed more important than others. In our variants, we change the probabilistic impact of links in accordance to their context/position on the page and measure the effects on the output of the PageRank algorithm. We compare the resulting rankings and those of existing systems with page-view-based rankings and provide statistics on the pairwise computed Spearman and Kendall rank correlations.

SUMMA: A Common API for Linked Data Entity Summaries

Linked Data knowledge sources such as DBpedia, Freebase, and Wikidata currently offer large amounts of factual data. As the amount of information that can be grasped by users is limited, data summaries are needed. If a summary relates to a specific entity we refer to it as entity summarization. Unfortunately, in many settings, the summaries of entities are tightly bound to user interfaces. This practice poses problems for efficient and objective comparison and evaluation. In this paper we focus on the question of how to make summaries exchangeable between multiple interfaces and multiple summarization services in order to facilitate evaluation and testing. We introduce SUMMA, an API definition that enables to decouple generation and presentation of summaries. It enables multiple consumers to retrieve summaries from multiple providers in a unified and lightweight way.

Big Data Analysis

The value of big data is predicated on the ability to detect trends and patterns and more generally to make sense of the large volumes of data that is often comprised of a heterogeneous mix of format, structure, and semantics. Big data analysis is the component of the big data value chain that focuses on transforming raw acquired data into a coherent usable resource suitable for analysis. Using a range of interviews with key stakeholders in small and large companies and academia, this chapter outlines key insights, state of the art, emerging trends, future requirements, and sectorial case studies for data analysis.

ELES: Combining Entity Linking and Entity Summarization

The automatic annotation of textual content with entities from a knowledge base is a well established field. Applications, such as DBpedia Spotlight and GATE enable to identify and disambiguate entities of text at high levels of accuracy. The output of such systems can be used in many different ways. One way is to show knowledge panels which provide a fact-based summary of an entity and provides further information as well as browsing options. Such fact-based summaries are produced by entity summarization systems.

Entity-centric Data Fusion on the Web

A lot of current web pages include structured data which can directly be processed and used. Search engines, in particular, gather that structured data and provide question answering capabilities over the integrated data with an entity-centric presentation of the results. Due to the decentralized nature of the web, multiple structured data sources can provide similar information about an entity. But data from different sources may involve different vocabularies and modeling granularities, which makes integration difficult. We present an approach that identifies similar entity-specific data across sources, independent of the vocabulary and data modeling choices. We apply our method along the scenario of a trustable knowledge panel, conduct experiments in which we identify and process entity data from web sources, and compare the output to a competing system. The results underline the advantages of the presented entity-centric data fusion approach.