Overview
Background
I am currently a lecturer at the University of Queensland, where I conduct research in the field of Information Retrieval. My research focuses on efficient and effective representations for large-scale search engines, including indexing, compression, and retrieval. I am also interested in understanding how to measure improvements in the end-to-end search pipeline, including system-oriented effectiveness measurements and user behaviour analysis. I have a broad interest in empirical experimentation, operating systems, data structures, and algorithms.
Previous Positions
- From February 2020 to January 2022 I worked as a postdoctoral research fellow on an ARC discovery project with Professor Alistair Moffat at the University of Melbourne.
- I completed my PhD at RMIT University under the guidance of Professor J. Shane Culpepper and Professor Falk Scholer.
Availability
- Dr Joel Mackenzie is:
- Available for supervision
- Media expert
Fields of research
Research interests
-
Efficient Data Structures and Algorithms
Inverted Indexes, Compression, Query Processing, Distributed Search, Parallel Processing, Data structures, Algorithms
-
Evaluating Search Systems
Empirical Experimentation, Metrics, Data Visualization, User Behaviour
Works
Search Professor Joel Mackenzie’s works on UQ eSpace
2024
Conference Publication
What do users really ask large language models? An initial log analysis of Google Bard interactions in the wild
Trippas, Johanne R., Al Lawati, Sara Fahad Dawood, Mackenzie, Joel and Gallagher, Luke (2024). What do users really ask large language models? An initial log analysis of Google Bard interactions in the wild. SIGIR '24: 47th International ACM SIGIR Conference on Research and Development in Information Retrieval, Washington, DC, United States, 14-18 July 2024. New York, NY, United States: ACM. doi: 10.1145/3626772.3657914
2024
Conference Publication
Revisiting document expansion and filtering for effective first-stage retrieval
Mansour, Watheq, Zhuang, Shengyao, Zuccon, Guido and Mackenzie, Joel (2024). Revisiting document expansion and filtering for effective first-stage retrieval. SIGIR '24, Washington, DC, United States, 14-18 July 2024. New York, NY, United States: ACM. doi: 10.1145/3626772.3657850
2024
Conference Publication
ReNeuIR at SIGIR 2024: The Third Workshop on Reaching Efficiency in Neural Information Retrieval
Fröbe, Maik, Mackenzie, Joel, Mitra, Bhaskar, Nardini, Franco Maria and Potthast, Martin (2024). ReNeuIR at SIGIR 2024: The Third Workshop on Reaching Efficiency in Neural Information Retrieval. SIGIR '24, Washington, DC, United States, 14 - 18 July 2024. New York, NY, United States: ACM. doi: 10.1145/3626772.3657994
2024
Journal Article
How much freedom does an effectiveness metric really have?
Moffat, Alistair and Mackenzie, Joel (2024). How much freedom does an effectiveness metric really have?. Journal of the Association for Information Science and Technology, 75 (6), 686-703. doi: 10.1002/asi.24874
2023
Conference Publication
Lossy compression options for dense index retention
Mackenzie, Joel and Moffat, Alistair (2023). Lossy compression options for dense index retention. 1st Annual International ACM SIGIR Conference on Research and Development in Information Retrieval in the Asia Pacific Region (SIGIR-AP), Beijing, Peoples Republic of China, 26-28 November 2023. New York, NY, United States: ACM. doi: 10.1145/3624918.3625316
2023
Conference Publication
Exploring the representation power of SPLADE models
Mackenzie, Joel, Zhuang, Shengyao and Zuccon, Guido (2023). Exploring the representation power of SPLADE models. ICTIR '23: The 2023 ACM SIGIR International Conference on the Theory of Information Retrieval, Taipei, Taiwan, 23 July 2023. New York, NY United States: Association for Computing Machinery. doi: 10.1145/3578337.3605129
2023
Conference Publication
ReNeuIR at SIGIR 2023: The Second Workshop on Reaching Efficiency in Neural Information Retrieval
Bruch, Sebastian, Mackenzie, Joel, Maistro, Maria and Nardini, Franco Maria (2023). ReNeuIR at SIGIR 2023: The Second Workshop on Reaching Efficiency in Neural Information Retrieval. 46th International ACM SIGIR Conference on Research and Development in Information Retrieval, Taipei, Taiwan, 23-27 July 2023. New York, NY, United States: ACM. doi: 10.1145/3539618.3591922
2023
Conference Publication
Profiling and visualizing dynamic pruning algorithms
Li, Zhixuan and Mackenzie, Joel (2023). Profiling and visualizing dynamic pruning algorithms. SIGIR 2023 - the 46th International ACM SIGIR Conference on Research and Development in Information Retrieval, Taipei, Taiwan, 23 - 27 July 2023. New York, NY United States: Association for Computing Machinery. doi: 10.1145/3539618.3591806
2023
Journal Article
Efficient immediate-access dynamic indexing
Moffat, Alistair and Mackenzie, Joel (2023). Efficient immediate-access dynamic indexing. Information Processing and Management, 60 (3) 103248, 1-24. doi: 10.1016/j.ipm.2022.103248
2023
Journal Article
Efficient document-at-a-time and score-at-a-time query evaluation for learned sparse representations
Mackenzie, Joel, Trotman, Andrew and Lin, Jimmy (2023). Efficient document-at-a-time and score-at-a-time query evaluation for learned sparse representations. ACM Transactions on Information Systems, 41 (4) 96. doi: 10.1145/3576922
2023
Conference Publication
Accelerating learned sparse indexes via term impact decomposition
MacKenzie, Joel, Mallia, Antonio, Moffat, Alistair and Petri, Matthias (2023). Accelerating learned sparse indexes via term impact decomposition. IIR 2023: Italian Information Retrieval Workshop, Pisa, Italy, 8-9 June 2023. Aachen, Germany: Rheinisch-Westfaelische Technische Hochschule Aachen Lehrstuhl Informatik.
2023
Conference Publication
Index-based batch query processing revisited
Mackenzie, Joel and Moffat, Alistair (2023). Index-based batch query processing revisited. 45th European Conference on Information Retrieval, ECIR 2023, Dublin, Ireland, 2-6 April 2023. Cham, Switzerland: Springer Nature Switzerland. doi: 10.1007/978-3-031-28241-6_6
2023
Other Outputs
CC-News-En
Mackenzie, Joel (2023). CC-News-En. The University of Queensland. (Dataset) doi: 10.48610/1dcb974
2022
Conference Publication
Greetings from the ADCS 2022 Chairs
Du, Jia Tina, Mackenzie, Joel, Nasim, Mehwish and Rybinski, Maciek (2022). Greetings from the ADCS 2022 Chairs. ADCS '22: Australasian Document Computing Symposium, Adelaide, SA, Australia, 5 - 16 December 2022. New York, NY, United States: Association for Computing Machinery. doi: 10.1145/3572960
2022
Conference Publication
Immediate-access indexing using space-efficient extensible arrays
Moffat, Alistair and Mackenzie, Joel (2022). Immediate-access indexing using space-efficient extensible arrays. ADCS '22: Australasian Document Computing Symposium, Adelaide, SA, Australia, 15-16 December 2022. New York, United States: Association for Computing Machinery. doi: 10.1145/3572960.3572984
2022
Conference Publication
Accelerating Learned Sparse Indexes Via Term Impact Decomposition
Mackenzie, Joel, Mallia, Antonio, Moffat, Alistair and Petri, Matthias (2022). Accelerating Learned Sparse Indexes Via Term Impact Decomposition. Conference on Empirical Methods in Natural Language Processing , Abu Dhabi, United Arab Emirates, 7-11 December 2022. Stroudsburg, PA United States: Association for Computational Linguistics.
2022
Conference Publication
IOQP: A simple Impact-Ordered Query Processor written in Rust
Mackenzie, Joel , Petri, Matthias and Gallagher, Luke (2022). IOQP: A simple Impact-Ordered Query Processor written in Rust. DESIRES 2022 – 3rd International Conference on Design of Experimental Search & Information REtrieval Systems, San Jose, CA, United States, 30-31 August 2022. Aachen, Germany: Rheinisch-Westfaelische Technische Hochschule Aachen.
2022
Conference Publication
A flexible framework for offline effectiveness metrics
Moffat, Alistair, MacKenzie, Joel, Thomas, Paul and Azzopardi, Leif (2022). A flexible framework for offline effectiveness metrics. 45th International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR), Madrid, Spain, 11-15 July 2022. New York, NY, United States: Association for Computing Machinery. doi: 10.1145/3477495.3531924
2022
Conference Publication
A common framework for exploring document-at-a-time and score-at-a-time retrieval methods
Trotman, Andrew, MacKenzie, Joel, Parameswaran, Pradeesh and Lin, Jimmy (2022). A common framework for exploring document-at-a-time and score-at-a-time retrieval methods. 45th International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR), Madrid, Spain, 11-15 July 2022. New York, NY, United States: Association for Computing Machinery. doi: 10.1145/3477495.3531657
2022
Conference Publication
Faster learned sparse retrieval with guided traversal
Mallia, Antonio, MacKenzie, Joel, Suel, Torsten and Tonellotto, Nicola (2022). Faster learned sparse retrieval with guided traversal. 45th International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR), Madrid, Spain, 11-15 July 2022. New York, NY, United States: Association for Computing Machinery. doi: 10.1145/3477495.3531774
Funding
Current funding
Supervision
Availability
- Dr Joel Mackenzie is:
- Available for supervision
Before you email them, read our advice on how to contact a supervisor.
Supervision history
Current supervision
-
Doctor Philosophy
Efficient Next-Generation Information Retrieval Systems
Principal Advisor
Other advisors: Professor Shane Culpepper
-
Master Philosophy
Search and Recommendation using Large Language Models
Associate Advisor
Other advisors: Professor Shane Culpepper
Completed supervision
-
2023
Doctor Philosophy
Teaching Pre-trained Language Models to Rank Effectively, Efficiently, and Robustly
Associate Advisor
Other advisors: Professor Guido Zuccon
Media
Enquiries
Contact Dr Joel Mackenzie directly for media enquiries about:
- Big Data
- Data Science
- Data Structures and Algorithms
- Information Retrieval
- Search Algorithms
- Search Engines
- Web Search
Need help?
For help with finding experts, story ideas and media enquiries, contact our Media team: