Distributed EDLSI, BM25, and Power Norm at TREC 2008
Bibliographic record
Abstract
Abstract : This paper describes our participation in the TREC Legal competition in 2008. Our first set of experiments involved the use of Latent Semantic Indexing (LSI) with a small number of dimensions, a technique we refer to as Essential Dimensions of Latent Semantic Indexing (EDLSI). Because the experimental dataset is large, we designed a distributed version of EDLSI to use for our submitted runs. We submitted two runs using distributed EDLSI, one with k = 10 and another with k = 41, where k is the dimensionality reduction parameter for LSI. We also submitted a traditional vector space baseline for comparison with the EDLSI results. This article describes our experimental design and the results of these experiments. We find that EDLSI clearly outperforms traditional vector space retrieval using a variety of TREC reporting metrics. We also describe experiments that were designed as a followup to our TREC Legal 2007 submission. These experiments test weighting and normalization schemes as well as techniques for relevance feedback. Our primary intent was to compare the BM25 weighting scheme to our power normalization technique. BM25 outperformed all of our other submissions on the competition metric (F1 at K) for both the ad hoc and relevance feedback tasks, but Power normalization outperformed BM25 in our ad hoc experiments when the 2007 metric (estimated recall at B) was used for comparison.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.024 | 0.033 |
| Meta-epidemiology (narrow) | 0.003 | 0.001 |
| Meta-epidemiology (broad) | 0.003 | 0.001 |
| Bibliometrics | 0.005 | 0.004 |
| Science and technology studies | 0.003 | 0.001 |
| Scholarly communication | 0.003 | 0.004 |
| Open science | 0.004 | 0.003 |
| Research integrity | 0.003 | 0.004 |
| Insufficient payload (model declined to judge) | 0.007 | 0.005 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".