iREAD: Interpretable REcognition and Automated Deconstruction of Semantics in Written English

iREAD is a learning aid that uses Machine Learning and Natural Language Processing techniques to deconstruct passages and identify literary devices and grammatical rules pertaining to Tense, Tone, Rhyme Schemes, and Metaphors, amongst others. The techniques used provide an insight to the rationale behind the deconstruction using: (1) an intuitive explanation through explication of the rules that underlie the grammatical constructs and (2) identifying the most relevant features for machine classification as a way to explain the outcome of models. We validate the algorithms that underlie the learning aid on manually annotated datasets from high school textbooks on English Grammar that are widely prescribed and referred to. The various components are delineated to three learning levels: easy (recognition of subject-predicate-object, tense and tone), intermediate (recognition of simile, alliteration and rhyme scheme) and advanced (recognition of metaphor and voice).

Published in the EDULEARN23 conference proceedings.

Required Libraries

To run the AllenNLP module, PyTorch is required (follow the installation steps on the website).

Run pip install -r requirements.txt once all the prerequisites are satisfied.

Execution

1. To start the authserver, navigate to webapp and run npm run dev on the terminal.
2. To start the client, navigate to webapp/client and run npm start on a new terminal.
3. To start the Flask server, navigate to src and run python server.py on another terminal.

Note: Refer to the Instructions file for detailed directions on how to execute the individual components.

Tech Specifications

• Frontend: React
• Backend: Python (Flask Server)
• Database for authentication: Firebase

Components Implemented

• Subject-Predicate-Object
• Tense
• Tone
• Alliterations
• Rhyme Scheme
• Simile
• Metaphor
• Voice

Citation

If you use this repository, please cite the following paper:

@InProceedings{SATISH2023IRE,
author    = {Satish, A. and Bhamare, A. and Hoskote, D. and Raj, V. and Srinivasa, G.},
title     = {IREAD: INTERPRETABLE RECOGNITION AND AUTOMATED DECONSTRUCTION OF SEMANTICS IN WRITTEN ENGLISH},
series    = {15th International Conference on Education and New Learning Technologies},
booktitle = {EDULEARN23 Proceedings},
isbn      = {978-84-09-52151-7},
issn      = {2340-1117},
doi       = {10.21125/edulearn.2023.0668},
url       = {https://doi.org/10.21125/edulearn.2023.0668},
publisher = {IATED},
location  = {Palma, Spain},
month     = {3-5 July, 2023},
year      = {2023},
pages     = {2266-2275}}