README.md

metadata

license: gpl-3.0
language:
  - en
pipeline_tag: text-classification
tags:
  - sbic
  - csr
datasets:
  - ia-nechaev/sbic-method2

sbic-method2

An updated version of Standard-Based Impact Classification (SBIC) method of CSR report analysis in accordance with GRI framework

Contents

1. Labeled dataset (150 International companies, 230 CSR GRI reports, 2017-2021 period, 57k paragraphs, 360k sentences)
2. Dataset for prediction (150 German PLC companies, 1.2k CSR reports, 2010-2021 period, 645k paragraphs) (full dataset available upon request)
3. Claculated text embeddings of both datasets
4. Script to predict the labels

Instructions on how to run the code below.

---

Multilabel Classification Steps

This code performs report similarity search using cosine similarity, K-Nearest Neighbor (KNN) algorithm, and Sigmoid activation function to classify reports based on embeddings.

Prerequisites

Ensure you have the following installed before running the script:

• Python 3.8+
• Required Python libraries (install using the command below)

pip install numpy pandas torch sentence-transformers scikit-learn

Input Files

Before running the script, make sure you have the following input files in the working directory:

1. Data Files:

   • labeled dataset: labeled.csv
   • dataset for prediction: prediction_demo.csv

2. Precomputed Embeddings:

   • labeled dataset: embeddings_labeled.pkl
   • dataset for prediction: embeddings_prediction.pkl

Running the Script

Run the script using the following command:

python script.py

Processing Steps

The script follows these main steps:

1. Load Data & Pretrained Embeddings
2. Perform Cosine Similarity Search: Finds the most relevant reports (sentences) using semantic_search from sentence-transformers.
3. Apply K-Nearest Neighbor (KNN) Algorithm: Selects top similar reports (sentences) and aggregates predictions.
4. Use Sigmoid Activation for Classification: Applies a threshold to generate final classification outputs.
5. Save Results: Exports df_results_0_50k.csv containing the processed data for the first 50k of records.

Output File

The processed results will be saved in: df_results_0_50k.csv

Execution Time

Execution time depends on the number of test samples and system resources. The script prints the total processing time upon completion.