nlp-modeler

Train NLP models: sentiment, classification, NER, topic modeling.

PublishedJun 16, 2026

Loading actions...

5 minBeginnerprompt4 files

Skill content

Main instructions and any bundled files for this skill.

markdown

Additional Files (3)

NLP Modeler

No hooks -- invoked via /nlp-classify, /nlp-ner, or /nlp-topics commands.

Capabilities

Sentiment Analysis

Lexicon-based (VADER, TextBlob polarity)
ML-based (logistic regression, SVM on TF-IDF features)
Transformer-based (fine-tuned BERT, DistilBERT)
Aspect-based sentiment extraction
Confidence calibration and threshold tuning

Text Classification

Traditional ML: Naive Bayes, Logistic Regression, SVM, Random Forest on TF-IDF
Deep learning: CNN text classifier, BiLSTM with attention
Transformer fine-tuning: BERT, DistilBERT, RoBERTa
Multi-label classification support
Class imbalance handling (oversampling, class weights, focal loss)

Named Entity Recognition

Rule-based (regex patterns, gazetteers)
spaCy NER models (built-in and custom training)
Transformer NER (token classification with BERT)
Custom entity type definition and training
Entity linking and disambiguation

Topic Modeling

LDA (Latent Dirichlet Allocation) with coherence optimization
NMF (Non-negative Matrix Factorization)
BERTopic (transformer-based topic modeling)
Dynamic topic modeling for temporal analysis
Topic labeling and interpretation

Model Evaluation

Classification: accuracy, precision, recall, F1 (macro/micro/weighted)
NER: entity-level precision, recall, F1, per-entity-type breakdown
Topic: coherence score (C_v, C_npmi), diversity, topic overlap
Confusion matrix and error analysis
Cross-validation with stratified splits

Report Bus

Write report using save_agent_report("nlp-modeler", {...}) with:

model type and architecture
training configuration (epochs, learning rate, batch size)
evaluation metrics (task-specific)
confusion matrix or entity-level breakdown
per-class/entity performance
recommendations for improvement

Contents

View Original Source

Related Skills

General

PromptBeginner5 minmarkdown

Untitled Skill

193

Jan 12, 2026

General

PromptBeginner5 minmarkdown

Frontend Typescript Linting.mdc

TypeScript and ESLint rules that MUST be followed when creating, modifying, or reviewing any file under apps/frontend/, including .ts, .tsx, .js, and .jsx files. Also apply when discussing frontend li...

160

Feb 15, 2026

General

PromptBeginner5 minmarkdown

2. Apply Deepthink Protocol (reason about dependencies

risks

126

Jan 15, 2026