Name: Crowdsourcing for entity linking and exploratory search for city councils
Start: 2018-06-29T16:00:00+02:00
End: 2018-06-29T18:00:00+02:00
Location: Academisch-cultureel centrum SPUI25

This Friday we'll have two talks followed by drinks. Our industrial speaker is Dyaa Albakour, a data scientist at Signal Media. He will talk about crowdsourcing for entity linking. Our academic speaker is Maarten Marx, Assistant Professor at the University of Amsterdam. He will talk about accessing city councils using exploratory search systems.

Program:

16:00-16:30 Dyaa Albakour

16:30-17:00 Maarten Marx

17:00-18:00 Drinks and snacks

Details:

Dyaa Albakour - Crowdsourcing at scale! The case of Entity Linking

In this presentation, we first review the current state-of-the-art for the EL task and make the case for using supervised learning approaches to tackle EL. These approaches require large amounts of labelled data, which represent a bottleneck for scaling them out to cover large numbers of entities. To mitigate this, we have developed a production-ready solution, powered by Active Learning, to collect high-quality labelled data at scale with Crowdsourcing. In particular, we will discuss the different steps and the challenges in tuning the design parameters of the crowdsourcing task. The design parameters include the qualification of the workers and UI features that help them complete the task. The tuning aims to limit the noise, reduce the cost and maximise the throughput of labelling whilst maintaining the quality of the resulting models for EL.

Bio: Dyaa works as a data scientist at a London-based company called Signal Media. In particular, Dyaa works with a team of scientists and developers to research and develop big-data text analytics services for a large-scale business intelligence platform. The platform supports reputation management, media coverage reporting, content marketing and other use cases. Prior to that, Dyaa was working as a Post-doctorate Researcher in the School of Computing Science at the University of Glasgow. He was working with Dr. Iadh Ounis and Dr. Craig Macdonald on the integrated Multimedia City Data (iMCD) project within Urban Big Data Centre. He also worked on the SMART FP7 EU project. Before coming to Glasgow, he was based in Colchester, Essex. Over there, he worked on a couple of research projects and completed his PhD in 2012 at the University of Essex under supervision of Dr. Udo Kruschwitz. His thesis is titled Adaptive Domain Modelling for Information Retrieval.

=====================================================================

Maarten Marx - Access to City Councils using Exploratory Search Systems

Anything that happens in the city, if important, is discussed in the city council. This makes the minutes of the city council a central hub connecting the city’s past, present, and future. Our ACCESS tools allow citizens, researchers, and policymakers to effectively explore any complex topic of the city, and contribute to government accountability and civic participation.

Bio: Maarten Marx (1964) obtained his master in political science (1990) and his PhD in mathematical logic (1995), both at the University of Amsterdam. He (co)-authored 3 books and more than 75 scientific articles. Since 2002 his main research topic is XML, in particular XPath dialects. In 2004 he won the ACM Principles of Database Systems best paper award for his Codd-completeness result for "Conditional XPath". His current research interest is integration of large amounts of semi-structured, text-centric, data. His work on the parliamentary proceedings was recognized with the XML Holland Award 2008 and the Dutch Data prize (awarded by DANS-KNAW) 2012.

Nikos Voskarides

SEA: Search Engines Amsterdam

Community & Environment

Science

Apache Solr

Technology

Information Architecture

Elasticsearch

Big Data

Web Technology

Software Engineering

Apache Lucene

SEO (Search Engine Optimization)

Crowdsourcing for entity linking and exploratory search for city councils

Academisch-cultureel centrum SPUI25

Delen

Crowdsourcing for entity linking and exploratory search for city councils

SEA: Search Engines Amsterdam

Details

Leden zijn ook geïnteresseerd in