simply_NLP

This repository contains a collection of projects developed during the Natural Language Processing (NLP) course. Each project addresses a specific topic and includes the necessary files for execution and understanding of the results.

Projects

1. Basicness

• Description: A project focused on analyzing and calculating the "basicness" of terms or concepts. A Random Forest model was used, starting from an annotated dataset and adding features derived from WordNet and a corpus of children's stories.
• Main Files:
  • basicness.ipynb: Main notebook for analysis.
  • basicness_f.py: Script for functions and logic.
• Results: Includes results in CSV format (fold1_results.csv).
• Requirements:

  pip install -r requirements.txt

2. content2form

• Description: This project performs onomasiological research, starting from definitions written by participants, and conducts a search in the WordNet graph following the "genus-differentia" principle.
• Main Files:
  • content2form.ipynb: Main notebook.
  • C2F_functions.py: Auxiliary functions script.
• Dataset: TLN-definitions-24.csv.
• Requirements:

  pip install -r requirements.txt

3. Defs

• Description: A project dedicated to the analysis of linguistic definitions. It aims to identify similarities between definitions provided by various participants.
• Main Files:
  • def.ipynb: Main notebook.
  • defs_f.py: Functions script.
• Dataset: TLN-definitions-24.csv.

4. FN

• Description: This project uses WordNet to disambiguate the various components of a frame in FrameNet.
• Main Files:
  • FN_algo.py: Core algorithms.
  • FN_main.ipynb: Main notebook.
  • JSON mapping files: mappings_hand_loris.json, mappings_hand_mattia.json, mappings_hand_simo.json.

5. ProfDanny_FINAL

• Description: This project aims to create a chatbot that acts as a professor, asking questions on NLP topics through a frame-based approach.
• Main Files:
  • dep.py: Core script.
  • frames.json: JSON file containing frames.
• Documentation: Includes a readme.txt file with additional details.

6. TweetLikeTrump

• Description: Project to generate texts similar to Trump tweets.
• Main Files:
  • ngrams.ipynb: Notebook on n-grams.
  • ngrams_lm.ipynb: Notebook for language models.
• Data: Stored in the data directory.

7. WSD

• Description: Project for Word Sense Disambiguation (WSD).
• Structure:
  • ConceptualSimilarity/: Implementations for conceptual similarity.
  • Lesk/: Algorithms based on the Lesk method.

General Requirements

Ensure you have installed:

• Python >= 3.8
• Libraries specified in the requirements.txt files in each project.

Repository Structure

.
├── Basicness/
├── content2form/
├── Defs/
├── FN/
├── ProfDanny_FINAL/
├── TweetLikeTrump/
├── WSD/
└── README.md