README.md

---
license: cc-by-nc-4.0
language:
- en
metrics:
- f1
pipeline_tag: text-classification
tags:
- transformers
- argument-mining
- opinion-mining
- information-extraction
- inference-extraction
- Twitter
widget:
- text: "Men shouldn’t be making laws about women’s bodies #abortion #Texas"
  example_title: "Statement"
- text: "’Bitter truth’: EU chief pours cold water on idea of Brits keeping EU citizenship after #Brexit HTTPURL via @USER"
  example_title: "Notification"
- text: "Opinion: As the draconian (and then some) abortion law takes effect in #Texas, this is not an idle question for millions of Americans. A slippery slope towards more like-minded Republican state legislatures to try to follow suit. #abortion #F24 HTTPURL"
  example_title: "Reason"
- text: "@USER Blah blah blah blah blah blah"
  example_title: "None"
- text: "republican men and karens make me sick"
  example_title: "Unlabeled 1"
- text: "No empire lives forever! Historical fact! GodWins! 🙏💪🇺🇲"
  example_title: "Unlabeled 2"
- text: "Further author information regarding registration and visa support letters will be sent to the authors soon. #CIKM2023"
  example_title: "Unlabeled 3"
- text: "Ummmmmm"
  example_title: "Unlabeled 4"
- text: "whoever says that The Last Jedi is a good movie is lying or trolling everyone"
  example_title: "Unlabeled 5"
- text: "I don’t think people realize how big this story is GBI Strategies, the group paid $11M+ by Biden PACs to harvest fraudulent voter registrations in *20 states*, may be the root source of Democrat election rigging @USER may have just exposed their entire fraud machine HTTPURL"
  example_tite: "Unlabeled 6"
---

# WRAP -- A TACO-based Classifier For Inference and Information-Driven Argument Mining on Twitter

Introducing WRAP, an advanced classification model built upon `AutoModelForSequenceClassification`, designed to identify tweets belonging to four 
distinct classes: Reason, Statement, Notification, and None of the [TACO dataset](https://anonymous.4open.science/r/TACO).
Designed specifically for extracting information and inferences from Twitter data, this specialized classification model utilizes 
[WRAPresentations](https://huggingface.co/TomatenMarc/WRAPresentations), from which WRAP acquires its name.
WRAPresentations is an advancement of the [BERTweet-base](https://huggingface.co/vinai/bertweet-base) architecture, whose embeddings were 
extended on augmented tweets using contrastive learning for better encoding inference and information in tweets.

## Class Semantics

The TACO framework revolves around the two key elements of an argument, as defined by the [Cambridge Dictionary](https://dictionary.cambridge.org).
It encodes *inference* as *a guess that you make or an opinion that you form based on the information that you have*, and it also leverages the
definition of *information* as *facts or details about a person, company, product, etc.*.

Taken together, WRAP can identify specific classes of tweets, where inferences and information can be aggregated in relation to these distinct
classes containing these components:

* *Statement*, which refers to unique cases where only the *inference* is presented as *something that someone says or writes officially, or an action
done to express an opinion*.
* *Reason*, which represents a full argument where the *inference* is based on direct *information* mentioned in the tweet, such as a source-reference
or quotation, and thus reveals the author’s motivation *to try to understand and to make judgments based on practical facts*.
* *Notification*, which refers to a tweet that limits itself to providing *information*, such as media channels promoting their latest articles.
* *None*, a tweet that provides neither *inference* nor *information*.

In its entirety, WRAP can classify the following hierarchy for tweets:

<div align="center">
  <img src="https://github.com/TomatenMarc/public-images/raw/main/Argument_Tree.svg" alt="Component Space" width="100%">
</div>

## Usage

Using this model becomes easy when you have `transformers` installed:

```python
pip install - U transformers
```

Then you can use the model to generate tweet classifications like this:

```python
from transformers import pipeline

pipe = pipeline("text-classification", model="TomatenMarc/WRAP")
prediction = pipe("Huggingface is awesome")

print(prediction)
```

<a href="https://anonymous.4open.science/r/TACO/notebooks/classifier_cv.ipynb">
    <blockquote style="border-left: 5px solid grey; background-color: #f0f5ff; padding: 10px;">
        Notice: The tweets need to undergo preprocessing before classification.
    </blockquote>
</a>

## Training

The final model underwent training using the entire shuffled ground truth dataset known as TACO, encompassing a total of 1734 tweets.
This dataset showcases the distribution of topics as: #abortion (25.9%), #brexit (29.0%), #got (11.0%), #lotrrop (12.1%), #squidgame (12.7%), and
#twittertakeover (9.3%). For training, we utilized [SimpleTransformers](https://simpletransformers.ai).

Additionally, the category and class distribution of the dataset TACO is as follows:

| Inference    | No-Inference |
|--------------|--------------|
| 865 (49.88%) | 869 (50.12%) |

| Information   | No-Information |
|---------------|----------------|
| 1081 (62.34%) | 653 (37.66%)   |

| Reason       | Statement    | Notification | None         |
|--------------|--------------|--------------|--------------|
| 581 (33.50%) | 284 (16.38%) | 500 (28.84%) | 369 (21.28%) |

<p>
    <blockquote style="border-left: 5px solid grey; background-color: #f0f5ff; padding: 10px;">
        Notice: Our training involved WRAP to forecast class predictions, where the categories (information/inference) represent class aggregations 
based on the inference or information component.
    </blockquote>
<p>

### Dataloader

```
"data_loader": {
    "type": "torch.utils.data.dataloader.DataLoader",
    "args": {
        "batch_size": 8,
        "sampler": "torch.utils.data.sampler.RandomSampler"
    }
}
```

Parameters of the fit()-Method:

```
{
    "epochs": 5,
    "max_grad_norm": 1,
    "optimizer_class": "<class 'torch.optim.adamw.AdamW'>",
    "optimizer_params": {
        "lr": 4e-05
    },
    "scheduler": "WarmupLinear",
    "warmup_steps": 66
}
```

## Evaluation

We applied a 6-fold (Closed-Topic) cross-validation method to demonstrate WRAP's optimal performance. This involved using the same dataset and parameters 
described in the *Training* section, where we trained on k-1 splits and made predictions using the kth split.

Additionally, we assessed its ability to generalize across the 6 topics (Cross-Topic) of TACO. Each of the k topics was utilized for testing, while 
the remaining k-1 topics were used for training purposes.

In total, the WRAP classifier performs as follows:

### Binary Classification Tasks

| Macro-F1     | Inference | Information | Multi-Class |
|--------------|-----------|-------------|-------------|
| Closed-Topic | 86.62%    | 86.30%      | 75.29%      |
| Cross-Topic  | 86.27%    | 84.90%      | 73.54%      |

### Multi-Class Classification Task

| Micro-F1     | Reason | Statement | Notification | None   |
|--------------|--------|-----------|--------------|--------|
| Closed-Topic | 78.14% | 60.96%    | 79.36%       | 82.72% |
| Cross-Topic  | 77.05% | 58.33%    | 78.45%       | 80.33% |

# Environmental Impact

- **Hardware Type:** A100 PCIe 40GB
- **Hours used:** 10 min
- **Cloud Provider:** [Google Cloud Platform](https://colab.research.google.com)
- **Compute Region:** [asia-southeast1](https://cloud.google.com/compute/docs/gpus/gpu-regions-zones?hl=en) (Singapore)
- **Carbon Emitted:** 0.02kg CO2

## Licensing

[WRAP](https://huggingface.co/TomatenMarc/WRAP) © 2023 is licensed under [CC BY-NC-SA 4.0](https://creativecommons.org/licenses/by-nc-sa/4.0/?ref=chooser-v1)