Add README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,44 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
## Jupyter Notebooks
|
| 2 |
+
GitHub link : [lihuicham/airbnb-helpfulness-classifier](https://github.com/lihuicham/airbnb-helpfulness-classifier)
|
| 3 |
+
|
| 4 |
+
Fine-tuning Python code in `finetuning.ipynb`
|
| 5 |
+
|
| 6 |
+
## Team Members (S001 - Synthetic Expert Team E) :
|
| 7 |
+
|
| 8 |
+
Li Hui Cham, Isaac Sparrow, Christopher Arraya, Nicholas Wong, Lei Zhang, Leonard Yang
|
| 9 |
+
|
| 10 |
+
## Description
|
| 11 |
+
This model is an AirBnB reviews helpfulness classifier. It can predict the helpfulness, from most helpful (A) to least helpful (C) of the reviews on AirBnB website.
|
| 12 |
+
|
| 13 |
+
## Pre-trained LLM
|
| 14 |
+
Our project fine-tuned [FacebookAI/roberta-base](https://huggingface.co/FacebookAI/roberta-base) for multi-class text (sequence) classification.
|
| 15 |
+
|
| 16 |
+
## Dataset
|
| 17 |
+
5000 samples are scraped from AirBnB website based on `listing_id` from this [Kaggle AirBnB Listings & Reviews dataset](https://www.kaggle.com/datasets/mysarahmadbhat/airbnb-listings-reviews).Samples were translated from French to English language.
|
| 18 |
+
|
| 19 |
+
Training Set : 4560 samples synthetically labelled by GPT-4 Turbo. Cost was approximately $60.
|
| 20 |
+
|
| 21 |
+
Test/Evaluation Set : 500 samples labelled manually by two groups (each group labelled 250 samples), majority votes applies. A scoring rubrics (shown below) is used for labelling.
|
| 22 |
+
|
| 23 |
+
## Training Details
|
| 24 |
+
```
|
| 25 |
+
hyperparameters = {'learning_rate': 3e-05,
|
| 26 |
+
'per_device_train_batch_size': 16,
|
| 27 |
+
'weight_decay': 1e-04,
|
| 28 |
+
'num_train_epochs': 4,
|
| 29 |
+
'warmup_steps': 500}
|
| 30 |
+
```
|
| 31 |
+
|
| 32 |
+
We trained our model on Colab Pro which costed us approximately 56 computing units.
|
| 33 |
+
|
| 34 |
+
## Slides
|
| 35 |
+
|
| 36 |
+
|
| 37 |
+
|
| 38 |
+
|
| 39 |
+
|
| 40 |
+
|
| 41 |
+
|
| 42 |
+
|
| 43 |
+
|
| 44 |
+
|