Anytime Ranking for Impact-Ordered Indexes

By Jimmy J. Lin, Andrew Trotman. 2015

63 citations recorded by the source corpus.

Subfields: Computer Networks and Communications, Information Systems, Information Retrieval & Search, Signal Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Networks and Communications

1. Matt Crane
2. Jimmy J. Lin
3. Hui Fang
4. Peilin Yang
5. Joel M. Mackenzie
6. J. Shane Culpepper
7. Andrew Trotman
8. Grant Ingersoll

In Information Systems

1. Jimmy J. Lin
2. Rodrigo Nogueira
3. Andrew Yates
4. David R. Cheriton
5. Jamie Callan
6. Joel M. Mackenzie
7. Peilin Yang
8. Hui Fang

In Information Retrieval & Search

1. Jimmy J. Lin
2. Rodrigo Nogueira
3. Andrew Yates
4. Jamie Callan
5. Zeynep Akkalyoncu Yilmaz
6. David R. Cheriton
7. Allan Hanbury
8. Sebastian Hofstätter

In Signal Processing

1. Jimmy J. Lin
2. Matt Crane
3. Joel M. Mackenzie
4. J. Shane Culpepper
5. Peilin Yang
6. Andrew Trotman
7. Hui Fang
8. Roi Blanco

Citations in the corpus

• Pretrained Transformers for Text Ranking: BERT and Beyond (2020, background): “i [Metzler and Croft, 2004, Metzler et al., 2004], Galago [Cartright et al., 2012], Terrier [Ounis et al., 2006, Macdonald et al., 2012], ATIRE [Trotman et al., 2012], Ivory [Lin et al., 2009], JASS [Lin and Trotman, 2015], JASSv2 [Trotman and Crane, 2019], MG4J [Boldi and Vigna, 2005], Wumpus,66 an”
• Anserini (2018, background): “An incomplete list includes Lemur/Indri [38, 39], Galago [15], Terrier [35, 42], ATIRE [51], Ivory [30], JASS [31], MG4J [12], [60, 61].”
• The Impact of Score Ties on Repeatability in Document Ranking (2018, background): “This relates to impact quantization [1], and JASS [6] is an example of a recent system that exploits approximate scoring for anytime ranking.”
• The Lucene for Information Access and Retrieval Research (LIARR) Workshop at SIGIR 2017 (2017, methodology): “An incomplete list includes SMART [18], Lemur/Indri [15, 16], Galago [8], Terrier [14, 17], ATIRE [20], Ivory [12], JASS [13], MG4J [6], Wumpus [21], and ZeŠair [25].”
• Query Driven Algorithm Selection in Early Stage Retrieval (2017, methodology): “The recommended heuristic value of ρ is 10% of the size of the collection [32], which is around 5 million for the ClueWeb09B collection.”
• Toward Reproducible Baselines: The Open-Source IR Reproducibility Challenge (2016, methodology): “JASS is a new, lightweight search engine built to explore score-at-a-time query evaluation on quantized indexes and the notion of “anytime” ranking functions [9].”
• Micro‐ and macro‐optimizations of SaaT search (2019, background): “5 , 6 First, we microscopically examine the cost of accessing member variables in C++ objects and show that there is essentially no overhead to it, leading to a design decision to object orient the implementation of our new parallel search engine.”
• A Comparison of Document-at-a-Time and Score-at-a-Time Query Evaluation (2017, methodology): “For JASS, Lin and Trotman [20] suggest a heuristic of setting ρ to be 10% of the collection size—these are shown by the red crosses on the JASS curves.”