krisztianbalog.com
temporal IR | Krisztian Balog
http://krisztianbalog.com/tag/temporal-ir
From Answering Engines to Task-completion Engines. Evaluating document filtering systems over time. June 12, 2015. July 15, 2015. Performance of three systems over time. Systems A and B degrade, while System C improves over time, but they all have the same average performance over the entire period. We express the change in system performance using the derivative of the fitted line (in orange) and compare performance at what we call the estimated end-point (the large orange dots). ICTIR 2016 paper online.
krisztianbalog.com
Evaluating document filtering systems over time | Krisztian Balog
http://krisztianbalog.com/2015-06-12/evaluating-document-filtering-systems-over-time
From Answering Engines to Task-completion Engines. Evaluating document filtering systems over time. June 12, 2015. July 15, 2015. Performance of three systems over time. Systems A and B degrade, while System C improves over time, but they all have the same average performance over the entire period. We express the change in system performance using the derivative of the fitted line (in orange) and compare performance at what we call the estimated end-point (the large orange dots). ICTIR 2016 paper online.
krisztianbalog.com
TREC KBA | Krisztian Balog
http://krisztianbalog.com/tag/trec-kba
From Answering Engines to Task-completion Engines. Evaluating document filtering systems over time. June 12, 2015. July 15, 2015. Performance of three systems over time. Systems A and B degrade, while System C improves over time, but they all have the same average performance over the entire period. We express the change in system performance using the derivative of the fitted line (in orange) and compare performance at what we call the estimated end-point (the large orange dots). ICTIR 2016 paper online.
krisztianbalog.com
Publications | Krisztian Balog
http://krisztianbalog.com/category/publications
From Answering Engines to Task-completion Engines. ICTIR 2016 paper online. July 21, 2016. 8220;Exploiting Entity Linking in Queries for Entity Retrieval,” an upcoming ICTIR 2016 paper by Faegheh Hasibi, Svein Erik Bratsberg, and myself is available. Online now, with resources coming soon. March 22, 2016. Last Sunday, Anne Schuth. And I gave a tutorial on Living Labs for Online Evaluation. The tutorial’s homepage. Contains all the slides and reference material. Today, Faegheh Hashibi. September 2, 2015.
diffeo.com
About | Diffeo
https://www.diffeo.com/about.html
WE IMPROVE HOW PEOPLE DIGEST COMPLEX INFORMATION. Diffeo is a knowledge discovery company that helps people organize the world's information by mining unbounded streams. Of Big Data and Big Text using entity coreference resolution. To enable entity-centric search. Of the Web and private content. Diffeo grew out of TREC KBA. Which we organized for NIST, and DARPA Memex. Open source tools and and. For named entity recognition in seventeen languages. This extends recent research results. Keywords alone lack...
krisztianbalog.com
evaluation | Krisztian Balog
http://krisztianbalog.com/tag/evaluation
From Answering Engines to Task-completion Engines. Evaluating document filtering systems over time. June 12, 2015. July 15, 2015. Performance of three systems over time. Systems A and B degrade, while System C improves over time, but they all have the same average performance over the entire period. We express the change in system performance using the derivative of the fitted line (in orange) and compare performance at what we call the estimated end-point (the large orange dots). ICTIR 2016 paper online.
streamcorpus.org
1. Overview of StreamCorpus software components — streamcorpus-pipeline 0.7.10.dev1 documentation
http://streamcorpus.org/sphinx-docs/overview.html
Streamcorpus-pipeline 0.7.10.dev1 documentation. 1 Overview of StreamCorpus software components. StreamCorpus provides a toolkit for processing massive streams of text and running natural language extractors on the text. StreamCorpus does not provide any extractors itself; it operates third-party extractors, such as Serif and Factorie, and unifies their output. StreamCorpus pipline can store documents in S3, Accumulo, or flat files. StreamCorpus is used extensively in TREC KBA. Embedded in StreamItem....