Defining Clusters by Topology Warping Features, an Interpretable Data Clustering Method
Bibliographic record
Abstract
Clustering is the task of dividing a data-set into different groups, called clusters, based on similarity. Despite being extensively studied, many state-of-the-art clustering algorithms lack interpretability. While it is useful to partition objects into clusters, it can be equally useful to understand why each cluster has been created. This motivates the desire for a clustering algorithm which can explain its partition. Our proposed method is a clustering process which seeks to explain which variables of a data-set are responsible for each cluster. We analyze the shape of the data, through the mathematical concept of a topological space. A space which is optimal for clustering is one which contains several disconnected islands, quantified as the number of connected components. Subsets of variables can then be selected and the resulting connected components of the topological space calculated. If this space is promising then the resulting disconnected region is the set of data-points which make up a cluster, and the subset of variables are what define it. We chose the variables to consider by grouping them via their correlation or through complex methods, e.g evolutionary algorithms. Our method provides a simple explanation since we can confidently assert that particular variables are why a data point is in a certain cluster. Alongside test data for comparison with existing algorithms, our methodology was applied to a community-based adolescents lipidomics dataset. Results on this dataset revealed 3 distinct clusters which can be explained by the distinct set of lipids that define them. Further optimization showed the existence of a cluster defined by a single lipid.
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 | 0.003 | 0.012 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.005 | 0.004 |
| Science and technology studies | 0.001 | 0.002 |
| Scholarly communication | 0.004 | 0.003 |
| Open science | 0.002 | 0.002 |
| Research integrity | 0.002 | 0.002 |
| Insufficient payload (model declined to judge) | 0.003 | 0.002 |
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".