How amishgodson/ettuthogai-poem-analysis works
This project classifies Tamil poem input as Ettuthogai or non-Ettuthogai, then analyzes accepted poems through tokenization, dictionary-based entity tagging, metadata matching, and meaning generation. Users can submit poems through the Flask interface or the command-line entry point. The README also describes corpus-based fallback validation, but the sampled runtime flow shows a pickled validator; its connection to the separate validator code is unverified. The graph follows the sampled orchestration and data reads; it does not assume an external model file is present in the tree.
Main entry points: main_logic.py Analysis orchestrator · module3_knowledge_mapping.py Knowledge mapping · app.py Flask interface · main.py CLI interface
Components 13
User entry
- Flask interface
app.py - CLI interface
main.py
Classification
- Analysis orchestrator
main_logic.py - Poem classifier
main_logic.py - Corpus validator
knowledge_base/extra_datas/ettuthogai_validator.py
Poem analysis
- Tamil preprocessing
modules/module1_preprocessing.py - Entity extraction
modules/module2_entity_extraction.py - Knowledge mapping
modules/module3_knowledge_mapping.py - Meaning generation
modules/module4_sentence_generation.py
Corpus data
- Entity dictionaries
data/ettuthogai/csv - Poem metadata corpus
data/ettuthogai/text_2
Other components
- Poem submitter
- Analysis result
Connections 15
- Poem submitter → Flask interface: submits poem
app.py - Flask interface → Analysis orchestrator: analyzes with
app.py - Poem submitter → CLI interface: enters poem
main.py - CLI interface → Analysis orchestrator: analyzes with
main.py - Analysis orchestrator → Poem classifier: classifies
main_logic.py - Analysis orchestrator → Tamil preprocessing: preprocesses
main_logic.py - Analysis orchestrator → Entity extraction: extracts entities
main_logic.py - Analysis orchestrator → Knowledge mapping: maps entities
main_logic.py - Analysis orchestrator → Knowledge mapping: reads metadata
main_logic.py - Analysis orchestrator → Knowledge mapping: counts entities
main_logic.py - Analysis orchestrator → Meaning generation: generates meaning
main_logic.py - Entity extraction → Entity dictionaries: loads dictionaries
modules/module2_entity_extraction.py - Knowledge mapping → Poem metadata corpus: searches poems
modules/module3_knowledge_mapping.py - Analysis orchestrator → Corpus validator: may validate
- Analysis orchestrator → Analysis result: returns analysis
main_logic.py