Deberta V3 Large Self Disclosure Detection

🚀 deberta-v3-large-自我披露檢測

該模型用於檢測句子中的自我披露信息（個人信息），這是一個多類別標記分類任務，類似於採用IOB2格式的命名實體識別（NER）。它能為保護個人隱私、識別敏感信息提供有力支持。

🚀 快速開始

此模型用於檢測句子中的自我披露信息（個人信息），是一個多類別標記分類任務，類似採用IOB2格式的NER。例如，句子 "I am 22 years old and ..." 的標籤為 "["B - Age", "I - Age", "I - Age", "I - Age", "I - Age", "O", ...]" 。

該模型能夠檢測以下17個類別："Age"（年齡）、"Age_Gender"（年齡與性別）、"Appearance"（外貌）、"Education"（教育程度）、"Family"（家庭情況）、"Finance"（財務狀況）、"Gender"（性別）、"Health"（健康狀況）、"Husband_BF"（丈夫/男友）、"Location"（地理位置）、"Mental_Health"（心理健康）、"Occupation"（職業）、"Pet"（寵物）、"Race_Nationality"（種族/國籍）、"Relationship_Status"（感情狀況）、"Sexual_Orientation"（性取向）、"Wife_GF"（妻子/女友）。

如需更多詳細信息，請閱讀論文：Reducing Privacy Risks in Online Self - Disclosures with Language Models 。

使用此模型意味著自動同意以下準則：

1. 僅將模型用於研究目的。
2. 未經作者同意，不得重新分發。
3. 任何使用此模型創建的衍生作品必須承認原作者。

✨ 主要特性

• 能夠精準檢測句子中17個類別的自我披露信息。
• 在相關評估指標上表現出色，部分跨度F1值達到65.71，優於提示GPT - 4（F1值為57.68）。

📦 安裝指南

文檔中未提及具體安裝步驟，故跳過該章節。

💻 使用示例

基礎用法

import torch
from torch.utils.data import DataLoader, Dataset

import datasets
from datasets import ClassLabel, load_dataset

from transformers import AutoModelForTokenClassification, AutoTokenizer, AutoConfig, DataCollatorForTokenClassification

model_path = "douy/deberta-v3-large-self-disclosure-detection"

config = AutoConfig.from_pretrained(model_path,)

tokenizer = AutoTokenizer.from_pretrained(model_path, use_fast=True,)

model = AutoModelForTokenClassification.from_pretrained(model_path,config=config,device_map="cuda:0").eval()

label2id = config.label2id
id2label = config.id2label


def tokenize_and_align_labels(words):
    tokenized_inputs = tokenizer(
                words,
                padding=False,
                is_split_into_words=True,
            )

    # we use ("O") for all the labels
    word_ids = tokenized_inputs.word_ids(0)
    previous_word_idx = None
    label_ids = []
    for word_idx in word_ids:
        # Special tokens have a word id that is None. We set the label to -100 so they are automatically
        # ignored in the loss function.
        if word_idx is None:
            label_ids.append(-100)
        # We set the label for the first token of each word.
        elif word_idx != previous_word_idx:
            label_ids.append(label2id["O"])
        # For the other tokens in a word, we set the label to -100
        else:
            label_ids.append(-100)
        previous_word_idx = word_idx
    tokenized_inputs["labels"] = label_ids
    return tokenized_inputs

class DisclosureDataset(Dataset):
    def __init__(self, inputs, tokenizer, tokenize_and_align_labels_function):
        self.inputs = inputs
        self.tokenizer = tokenizer
        self.tokenize_and_align_labels_function = tokenize_and_align_labels_function

    def __len__(self):
        return len(self.inputs)

    def __getitem__(self, idx):
        words = self.inputs[idx]
        tokenized_inputs = self.tokenize_and_align_labels_function(words)
        return tokenized_inputs
    
    
sentences = [
    "I am a 23-year-old who is currently going through the last leg of undergraduate school.",
    "My husband and I live in US.",
]

inputs = [sentence.split() for sentence in sentences]

data_collator = DataCollatorForTokenClassification(tokenizer)

dataset = DisclosureDataset(inputs, tokenizer, tokenize_and_align_labels)

dataloader = DataLoader(dataset, collate_fn=data_collator, batch_size=2)

total_predictions = []
for step, batch in enumerate(dataloader):
    batch = {k: v.to(model.device) for k, v in batch.items()}
    with torch.inference_mode():
        outputs = model(**batch)
    predictions = outputs.logits.argmax(-1)
    labels = batch["labels"]

    predictions = predictions.cpu().tolist()
    labels = labels.cpu().tolist()

    true_predictions = []
    for i, label in enumerate(labels):
        true_pred = []
        for j, m in enumerate(label):
            if m != -100:
                true_pred.append(id2label[predictions[i][j]])
        true_predictions.append(true_pred)
    total_predictions.extend(true_predictions)
    

for word, pred in zip(inputs, total_predictions):
    for w, p in zip(word, pred):
        print(w, p)

📚 詳細文檔

模型描述

評估

該模型的部分跨度F1值達到65.71，優於提示GPT - 4（F1值為57.68）。有關每個類別的詳細性能，請參閱論文。

🔧 技術細節

文檔中未提及具體技術細節內容，故跳過該章節。

📄 許可證

該模型使用的許可證為知識共享署名 - 非商業性使用許可協議（Creative Commons Attribution - NonCommercial）。

📚 引用

@article{dou2023reducing,
  title={Reducing Privacy Risks in Online Self-Disclosures with Language Models},
  author={Dou, Yao and Krsek, Isadora and Naous, Tarek and Kabra, Anubha and Das, Sauvik and Ritter, Alan and Xu, Wei},
  journal={arXiv preprint arXiv:2311.09538},
  year={2023}
}