地球资源数据云——数据资源详情
虚假与真实职位发布（合成 NLP 数据集）
该数据集《Fake vs Real Job Postings (Synthetic NLP Dataset)》主要用于二分类任务，数据形态以文本为主，应用场景偏向文本内容分析。 题目说明：Fake vs Real Job Postings 任务类型：文本二分类。 建议流程：先做文本清洗与分词，再比较 TF - IDF+线性模型 与 预训练语言模型。 评估建议：使用分层切分或交叉验证，优先关注 F1、Recall、AUC 等分类指标。 可用文件：fake_real_job_postings_3000x25.csv。 This dataset contains 3,000 synthetic job postings designed to simulate realistic patterns found in legitimate and fraudulent online job advertisements. It is intended for Natural Language Processing (NLP) and text classification tasks, particularly for detecting fake or scam job postings. Each row represents a single job posting and includes a mix of: Free - text fields (job description, requirements, company profile) Categorical attributes (job title, industry, employment type) Numerical features (experience years, text length)
摘要概览
该数据集《Fake vs Real Job Postings (Synthetic NLP Dataset)》主要用于二分类任务，数据形态以文本为主，应用场景偏向文本内容分析。 题目说明：Fake vs Real Job Postings
任务类型：文本二分类。
建议流程：先做文本清洗与分词，再比较 TF - IDF+线性模型 与 预训练语言模型。
评估建议：使用分层切分或交叉验证，优先关注 F1、Recall、AUC 等分类指标。
可用文件：fake_real_job_postings_3000x25.csv。
This dataset contains 3,000 synthetic job postings designed to simulate realistic patterns found in legitimate and fraudulent online job advertisements. It is intended for Natural Language Processing (NLP) and text classification tasks, particularly for detecting fake or scam job postings.
Each row represents a single job posting and includes a mix of:
Free - text fields (job description, requirements, company profile)
Categorical attributes (job title, industry, employment type)
Numerical features (experience years, text length)