Big Data Clustering: Models and Applications
Bibliographic record
Abstract
This thesis presents frameworks for data clustering on big datasets that can arise in different real-world applications. The main contributions of this thesis can be divided into the following four areas of data clustering. Correlation clustering is a well-known problem that appears in different scientific areas with various names that identify clusters when qualitative information about objects' mutual similarities or dissimilarities is given. The first contribution of this thesis is to present a unified discussion on the cross-disciplinary taxonomy-based literature review, bibliometric analysis, literature gaps and dominant research topics related to this problem. As the second contribution, this thesis presents the concept of a common-knowledge network and a heuristic algorithm for clustering editing to identify authors' communities in a research institution. Furthermore, several analyses, such as the dominant research topic and collaboration incident corresponding to each identified research community, are proposed in this thesis to investigate multidisciplinary research activities in research institutions. The third contribution constitutes a framework for user-generated short-text classification based on identified line-item categories. The line-item identification phase uses cograph editing (CoE)-based clustering on keywords network formulated from short-texts. An integer linear programming formulation for CoE on weighted networks and a corresponding heuristic algorithm to identify clusters in large-scale networks are also proposed. The framework has been applied to categorize invoices for a subscription-based invoicing and accounting company. An augmented artificial intelligence (AI) hybrid fraud detection framework in the presence of minimal labelled data sets. This framework uses unsupervised clustering, a supervised classifier, red-flag prioritization, and augmented AI processes. Finally, this thesis outlines an application of this framework to identify fraudulent users in an invoicing platform.
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 | 0.007 | 0.022 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.002 |
| Bibliometrics | 0.005 | 0.008 |
| Science and technology studies | 0.002 | 0.003 |
| Scholarly communication | 0.006 | 0.007 |
| Open science | 0.004 | 0.004 |
| Research integrity | 0.004 | 0.003 |
| Insufficient payload (model declined to judge) | 0.004 | 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".