MétaCan
Menu
Back to cohort

Improving Topic Quality with Interactive Beta-Liouville Mixture Allocation Model

2022· article· en· W4318606113 on OpenAlex
Kamal Maanicshah, Manar Amayri, Nizar Bouguila

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

Venue2022 IEEE Symposium Series on Computational Intelligence (SSCI) · 2022
Typearticle
Languageen
FieldComputer Science
TopicTopic Modeling
Canadian institutionsConcordia University
Fundersnot available
KeywordsLatent Dirichlet allocationTopic modelComputer scienceInferenceCluster analysisCategorizationArtificial intelligenceDirichlet distributionNatural language processingTask (project management)Machine learningSet (abstract data type)Quality (philosophy)Hierarchical Dirichlet processBETA (programming language)Mathematics

Abstract

fetched live from OpenAlex

One of the major tasks in natural language processing is to categorize texts into different categories. Topic models are an important set of tools for categorizing texts and so are mixture models since both models learn patterns from data in an unsupervised manner. The introduction of latent Dirichlet allocation (LDA) triggered a lot of research in this domain. Recent research investigates the use of distributions other than Dirichlet for the topic proportions in LDA especially generalized Dirichlet and Beta-Liouville distributions in addition to adding useful attributes specific to the task at hand. Improving the quality of topics extracted from these models is important for accurate inference and unsupervised language tasks. Owing to this cause, in this paper, we propose interactive Beta-Liouville mixture allocation (iBLMA) model which combines the clustering capabilities of mixture models with interactive learning which helps the user modify the topic weights of irrelevant words within the topic. We show the efficiency of our model with experiments on two different text datasets.

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)0.001
metaresearch head score (Gemma)0.000
Version: codex-gemma-dda1882f352aValidation status: machine_predicted_unvalidated
Candidate categoriesMeta-epidemiology (narrow)
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Simulation or modeling · Consensus signal: Simulation or modeling
GenreCandidate signal: Methods · Consensus signal: none
Teacher disagreement score0.899
Threshold uncertainty score1.000

Codex and Gemma teacher scores by category

CategoryCodexGemma
Metaresearch0.0010.000
Meta-epidemiology (narrow)0.0000.000
Meta-epidemiology (broad)0.0000.000
Bibliometrics0.0000.001
Science and technology studies0.0010.000
Scholarly communication0.0000.001
Open science0.0010.000
Research integrity0.0000.001
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 head0.026
GPT teacher head0.281
Teacher spread0.254 · 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