E-commerce_Platform_Screen_Recording_Dataset

# E-commerce Platform Screen Recording Dataset
*This dataset contains high-quality screen recordings of user interactions on an e-commerce platform. It has been carefully curated, cleaned, and anonymized to ensure accuracy, completeness, and compliance with privacy standards, making it suitable for UI/UX research, AI training, and behavior analysis.*

## Contact
For queries or collaborations related to this dataset, contact:
- [e-mail]
- [e-mail]

## Supported Tasks

- **Task Categories**:
- Video Classification
- Video-to-Text Generation

- **Supported Tasks**:
- Automatic captioning of screen recordings
- User journey segmentation and analysis
- Action recognition (e.g., search, click, checkout)
- Usability and UX pattern analysis
- Clickstream behavior prediction
- E-commerce workflow automation

## Languages

- **Primary Language**: English (UI, product listings, and user interactions)

## Dataset Creation

### Curation Rationale
This dataset was created to advance research in user interaction modeling, e-commerce usability analysis, and multimodal AI systems. It enables training models that can understand complex UI workflows, user intent, and purchasing behavior in digital marketplaces.

### Source Data
- **Contributors**: Collected from simulated user sessions on an e-commerce platform
- **Collection Process**: Screen recordings of diverse user activities including browsing, searching, adding items to cart, and completing purchases. Personally identifiable information (PII) is anonymized or removed.

### Other Known Limitations
- **Size**: Dataset may not cover all e-commerce workflows (e.g., returns, customer support)
- **Bias**: Recorded sessions may overrepresent certain product categories or browsing behaviors
- **Platform-Specificity**: May not generalize to all e-commerce platforms or mobile-first interfaces

## Intended Uses

### ✅ Direct Use
- Training AI models for action recognition in screen recordings
- Research in human-computer interaction (HCI) and usability testing
- Automatic summarization of user journeys
- Academic research in e-commerce UX and behavioral analysis

### ❌ Out-of-Scope Use
- Tracking or surveillance of real users without consent
- Commercial exploitation without attribution or ethical clearance
- Inference of personal or sensitive information from recordings
- Real-time monitoring or decision-making without validation

## License

CC BY 4.0

# 电商平台屏幕录制数据集
*本数据集收录电商平台上用户交互场景的高质量屏幕录制视频。该数据集经过严格遴选、清理与匿名化处理，以保障数据准确性、完整性及隐私合规性，可适用于用户界面/用户体验（UI/UX）研究、人工智能训练及行为分析等多类场景。*

## 联系方式
若您对本数据集存在疑问或合作需求，请联系：
- [e-mail]
- [e-mail]

## 支持任务
- **任务类别**：
- 视频分类
- 视频到文本生成
- **支持的具体任务**：
- 屏幕录制视频自动字幕生成
- 用户旅程分割与分析
- 行为识别（如搜索、点击、结账操作）
- 可用性与用户体验模式分析
- 点击流行为预测
- 电商工作流自动化

## 语言
- **主要语言**：英语（覆盖界面文本、商品列表及用户交互内容）

## 数据集构建
### 遴选依据
本数据集旨在推动用户交互建模、电商可用性分析及多模态人工智能系统领域的研究发展，可用于训练能够理解数字市场中复杂用户界面工作流、用户意图与购买行为的模型。

### 源数据
- **贡献主体**：数据采集自某电商平台的模拟用户会话
- **采集流程**：录制涵盖浏览、搜索、加购、完成购买等多样化用户活动的屏幕视频。所有个人可识别信息（Personally Identifiable Information，PII）均已完成匿名化或移除处理。

### 其他已知局限性
- **覆盖范围**：本数据集未覆盖全部电商工作流（如退货、客户支持流程等）
- **数据偏差**：录制的会话可能过度侧重特定商品类别或浏览行为模式
- **平台特异性**：本数据集可能无法适配所有电商平台或移动端优先的界面设计

## 预期用途
### ✅ 直接使用场景
- 训练用于屏幕录制视频行为识别的人工智能模型
- 人机交互（Human-Computer Interaction，HCI）与可用性测试相关研究
- 用户旅程自动摘要生成
- 电商用户体验与行为分析领域的学术研究

### ❌ 非适用使用场景
- 未经用户同意对真实用户进行追踪或监控
- 未获得授权或伦理许可的商业开发与利用
- 从录制视频中推断个人或敏感信息
- 未经过验证的实时监控或决策应用

## 许可协议
CC BY 4.0