MétaCan
Menu
Back to cohort
Record W49807964

A simple feature selection method for text classification

2001· article· en· W49807964 on OpenAlex

Why this work is in the frame

A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.

affAt least one author lists a Canadian institution in the pinned OpenAlex snapshot.

Bibliographic record

VenueInternational Joint Conference on Artificial Intelligence · 2001
Typearticle
Languageen
FieldComputer Science
TopicText and Document Classification Technologies
Canadian institutionsUniversité Laval
Fundersnot available
KeywordsFeature selectionComputer scienceInformation gainFeature (linguistics)Simple (philosophy)Set (abstract data type)Artificial intelligenceSelection (genetic algorithm)Pattern recognition (psychology)Data miningFeature extractionMachine learning
DOInot available

Abstract

fetched live from OpenAlex

In text classification most techniques use bag-of-words to represent documents. The main problem is to identify what words are best suited to classify the documents in such a way as to discriminate between them. Feature selection techniques are then needed to identify these words. The feature selection method presented in this paper is rather simple and computationally efficient. It combines a well known feature selection criterion, the information gain, and a new algorithm that selects and adds a feature to a bag-of-words if it does not occur too often with the features already in a small set composed of the best features selected so far for their high information gain. In brief, it tries to avoid considering features whose discrimination capability is sufficiently covered by already selected features, reducing in size the set of the features used to characterize the document set. This paper presents this feature selection method and its results, and how we have predetermined some of its parameters through experimentation.

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.

Full frame distilled prediction

Teacher imitation

Not 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.

metaresearch head score (Codex)0.000
metaresearch head score (Gemma)0.000
Version: codex-gemma-dda1882f352aValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Theoretical or conceptual · Consensus signal: none
GenreCandidate signal: Methods · Consensus signal: none
Teacher disagreement score0.889
Threshold uncertainty score0.869

Codex and Gemma teacher scores by category

CategoryCodexGemma
Metaresearch0.0000.000
Meta-epidemiology (narrow)0.0000.000
Meta-epidemiology (broad)0.0000.000
Bibliometrics0.0000.001
Science and technology studies0.0000.000
Scholarly communication0.0010.001
Open science0.0010.000
Research integrity0.0000.000
Insufficient payload (model declined to judge)0.0000.000

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.

Opus teacher head0.151
GPT teacher head0.384
Teacher spread0.232 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it