RSenter: Tool for Topics and Terms Extraction from Unstructured Data Debris
Bibliographic record
Abstract
There is enormous volume of user generated content (data) today in open source repositories, online social networks, and so on that enterprises can feed on to enhance product and services delivery. Apart from the open source data, enterprises are also generating a lot of data in-house since modern business requirements are shifting from paper-base to digital records. The major setback however is that, the data is unstructured in the sense that it is in heterogeneous formats (different file types including multimedia files), it is schema less, and it is scattered on multiple sources. This condition makes knowledge discovery (a.k.a. data mining) very challenging. Previous studies have proposed the hierarchical clustering methodology since it enhances human readability and provides clear dependency structure through topics, term and document organization. But, the methodology can be resource intensive and time consuming. Our work investigates the methodology and proposes a tool called RSenter that searches based on parallelization, random walk (or linear search), pessimistic search, and optimistic search in order to generate the hierarchical structure in real time within a search space. Currently, RSenter can search through NoSQL databases and HTML documents and traverse through all the links that are connected to that HTML to the nth depth, extracting the entire user specified elements (topics and terms). Further, the tool can search through an entire repository and organize the files in a hierarchical structure regardless of the file formats.
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 | 0.002 | 0.009 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.011 | 0.007 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.003 | 0.003 |
| Open science | 0.002 | 0.002 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.021 | 0.014 |
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".