University of Leipzig AKSW Homepage | Blog |

Archiv für die Kategorie 'PhD topic'

AKSW Colloquium: Edgard Marx and Tommaso Soru on Monday, February 23, 3.00 p.m.

February 19, 2015 - 10:53 pm by TommasoSoru - No comments »

On Monday, 23rd of February 2015, Edgard Marx will introduce Smart, a search engine designed over the Semantic Search paradigm; subsequently, Tommaso Soru will present ROCKER, a refinement operator approach for key discovery.

Abstract – Smart

Since the conception of the Web, search engines play a key role in making content available. However, retrieving of the desire information is still significantly challenging. Semantic Search systems are a natural evolution of the traditional search engines. They promise more accurate interpretation by understanding the contextual meaning of the user query. In this talk, we will introduce our audience to Smart, a search engine designed over the Semantic Search paradigm. Smart incorporates two of our currently designed approaches of dealing with the problem of Information Retrieval, as well as a novel interface paradigm. Moreover, we will present some of the former, as well as more recent state-of-the-art approaches used by the industry – for instance by Yahoo!, Google and Facebook.

Abstract – ROCKER

As within the typical entity-relationship model, unique and composite keys are of central importance also when their concept is applied on the Linked Data paradigm. They can provide help in manifold areas, such as entity search, question answering, data integration and link discovery. However, the current state of the art does not count approaches able to scale while relying on a correct definition of key. We thus present a refinement-operator-based approach dubbed ROCKER, which has shown to scale to big datasets with respect to the run time and the memory consumption. ROCKER will be officially introduced at the 24th International Conference on World Wide Web.

Tommaso Soru, Edgard Marx, and Axel-Cyrille Ngonga Ngomo, “ROCKER – A Refinement Operator for Key Discovery”. [PDF]

AKSW Colloquium: Konrad Höffner and Michael Röder on Monday, February 16, 3.00 p.m.

February 16, 2015 - 1:45 pm by KonradHoeffner - No comments »

CubeQA—Question Answering on Statistical Linked Data by Konrad Höffner

Abstract

Question answering systems provide intuitive access to data by translating natural language queries into SPARQL, which is the native query language of RDF knowledge bases. Statistical data, however, is structurally very different from other data and cannot be queried using existing approaches. Building upon a question corpus established in previous work, we created a benchmark for evaluating questions on statistical Linked Data in order to evaluate statistical question answering algorithms and to stimulate further research. Furthermore, we designed a question answering algorithm for statistical data, which covers a wide range of question types. To our knowledge, this is the first question answering approach for statistical RDF data and could open up a new research area.
See also the paper (preprint, under review) and the slides.

News from the WSDM 2015 by Michael Röder

Abstract

The WSDM conference is one of the major conferences for Web Search and Data Mining. Michael Röder was attending this years WSDM conference in Shanghai and wants to present a short overview over the conference topics. After that, he wants to take a closer look at FEL – an entity linking approach for search queries peresented at the conference.

About the AKSW Colloquium

This event is part of a series of events about Semantic Web technology. Please see http://wiki.aksw.org/Colloquium for further information about previous and future events. As always, Bachelor and Master students are able to get points for attendance and there is complimentary coffee and cake after the session.

AKSW Colloquium: Ricardo Usbeck and Ivan Ermilov on Monday, February 2, 3.00 p.m.

February 2, 2015 - 12:00 pm by Ricardo Usbeck - No comments »

GERBIL – General Entity Annotation Benchmark Framework by Ricardo Usbeck

Abstract

The need to bridge between the unstructured data on the document Web and the structured data on the Data Web has led to the development of a considerable number of annotation tools. Those tools are hard to compare since published results are calculated on diverse datasets and measured in different units.

We present GERBIL, a general entity annotation system based on the BAT-Framework. GERBIL offers an easy-to-use web-based platform for the agile comparison of annotators using multiple datasets and uniform measuring approaches. To add a tool to GERBIL, all the end user has to do is to provide a URL to a REST interface to its tool which abides by a given specification. The integration and benchmarking of the tool against user-specified datasets is then carried out automatically by the GERBIL platform. Currently, out platform provides results for 9 annotators and 11 datasets with more coming. Internally, GERBIL is based on the Natural Language Programming Interchange Format (NIF) and provide Java classes for implementing APIs for datasets and annotators to NIF. For the paper see here.

Towards Efficient and Effective Semantic Table Interpretation by Ziqi Zhang presented by Ivan Ermilov

Abstract

Ivan will present a paper that describes TableMiner by Ziqi Zhang, the first semantic Table Interpretation method that adopts an incremental, mutually recursive and bootstrapping learning approach seeded by automatically selected ‘partial’ data from a table. TableMiner labels columns containing named entity mentions with semantic concepts that best describe data in columns, and disambiguates entity content cells in these columns. TableMiner is able to use various types of contextual information outside tables for Table Interpretation, including semantic markups (e.g., RDFa/microdata annotations) that to the best of our knowledge, have never been used in Natural Language Processing tasks. Evaluation on two datasets shows that compared to two baselines, TableMiner consistently obtains the best performance. In the classification task, it achieves significant improvements of between 0.08 and 0.38 F1 depending on different baseline methods; in the disambiguation task, it outperforms both baselines by between 0.19 and 0.37 in Precision on one dataset, and between 0.02 and 0.03 F1 on the other dataset. Observation also shows that the bootstrapping learning approach adopted by TableMiner can potentially deliver computational savings of between 24 and 60% against classic methods that ‘exhaustively’ processes the entire table content to build features for interpretation.

About the AKSW Colloquium

This event is part of a series of events about Semantic Web technology. Please see http://wiki.aksw.org/Colloquium for further information about previous and future events. As always, Bachelor and Master students are able to get points for attendance and there is complimentary coffee and cake after the session.

AKSW Colloquium “Combination of Topic Modeling and Semantic Web” on Monday, June 30

June 26, 2014 - 9:12 am by KonradHoeffner - No comments »

Combination of Topic Modeling and Semantic Web

On Monday, June 30, at 3.00 p.m. in room Paulinum 702, Michael Röder will present his yearly PhD progress report “Combination of Topic Modeling and Semantic Web”. The presentation addresses the usage of Topic Modeling in the area of Semantic Web. We will focus a use case in which topic models shall be used to recommend similar RDF datasets for a given dataset.

About the AKSW Colloquium

This event is part of a series of events about Semantic Web technology. Please see http://wiki.aksw.org/Colloquium for further information about previous and future events. As always, Bachelor and Master students are able to get points for attendance and there is complimentary coffee and cake after the session.

AKSW Colloquium PhD student talks of Martin Brümmer and Ricardo Usbeck on Monday, May 19, 3.00 p.m.

May 16, 2014 - 9:23 am by KonradHoeffner - No comments »

“Exploitation scenarios for Wikipedia abstracts” by Martin Brümmer

Martin Brümmer has started as a researcher at AKSW in Dec. 2013. He is a contributor to the NLP2RDF and the DBpedia Project and was co-chair of the Multilingual Linked Data for Enterprises (MLODE) 2012 workshop. He contributed to the development of the Linguistic Linked Open Data Cloud with conversions of diverse linguistic datasets into RDF. He is also maintaining an RDF-backed database of sidejobs done by the members of the German parliament. His research focus is on Linguistic Linked Open Data, NLP in the Semantic Web and Open Government Data.

Abstract

In the context of the DBpedia project, most of the structured data contained in the Wikipedia has already been or is in the process of being extracted. However, the unstructured text containing most enzyclopaedic knowledge is largely unexplored. The talk will try to define some exploitation scenarios of Wikipedia abstracts with the goal of exploring directions for future research.

“Combining Linked Data and Statistical Information Retrieval” by Ricardo Usbeck

Abstract

Being a part of the Information Age, users are challenged with a tremendously growing amount of Web data which generates a need for more sophisticated information retrieval systems. Linked Data provides necessary procedures to augment the highly unstructured Web with suitable metadata in order to leverage search quality and user experience. In this talk, we will outline an approach for creating a web-scale, precise and efficient information system capable of understanding keyword, entity and natural language queries. By using Semantic Web methods the doctoral work will present how the underlying knowledge is created and elaborated searches can be performed on top.

About the AKSW Colloquium

This event is part of a series of events about Semantic Web technology. Please see http://wiki.aksw.org/Colloquium for further information about previous and future events. As always, Bachelor and Master students are able to get points for attendance and there is complimentary coffee and cake after the session.