Outliers and data mining : finding exceptions in data
Bibliographic record
Abstract
Our thesis is that we can efficiently identify meaningful outliers in large, multidimensional datasets. In particular, we introduce and study the notion and utility of distance-based outliers (DB-outliers). First, we focus on outlier identification, and present nested-loop, index-based, and cell-based algorithms. For datasets that are mainly disk-resident, we present another version of the cell-based algorithm that guarantees at most 3 passes over a dataset. We provide experimental results showing that these cell-based algorithms are by far the best for 4 or fewer dimensions. Second, we focus on the computation of intensional knowledge, that is, we provide a description or an explanation of why an identified outlier is exceptional. We provide an algorithm to compute the "strongest" outliers in a dataset (i.e., outliers that are dominant for certain attributes or dimensions). Our notion of strongest outliers allows significant pruning to take place in the search tree. We also define the notion of "weak" outliers. With respect to the computation of intensional knowledge, we develop naive, semi-naive, and I/O optimized algorithms. The latter class of algorithms intelligently schedules I/O's, and achieves optimization via page sharing. Third, we focus on robust space transformations to achieve meaningful results when mining Ac-D datasets for Z?.B-outliers. Robust estimation is used to: (a) account for differences among attributes in scale, variability, and correlation, (b) account for the effects of outliers in the data, and (c) prevent undesirable masking and flooding during the search for outliers. We propose using a robust space transformation called the Donoho-Stahel estimator (DSE), and we show key properties of the DSE. Of particular importance to data mining applications involving large or dynamic datasets is the stability property, which says that in spite of frequent updates, the estimator does not: (a) change much, (b) lose its usefulness, or (c) require re-computation. We develop randomized algorithms and evaluate how well they perform empirically. The novel algorithm that we develop is called the Hybrid-Random algorithm, which can significantly outperform the other DSE algorithms for moderate-high levels of recall. Experimental results using real-world data are included to show the utility and efficiency of Di>-outiiers.
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 distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.002 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.152 | 0.000 |
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 teacher head, 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".