Datasets for "Designing CAPTCHA Systems with Reinforcement Learning for Adaptive Defense"

This repository contains the datasets used in the study "Designing CAPTCHA Systems with Reinforcement Learning for Adaptive Defense." The repository includes: 1. Human interaction datasets collected during CAPTCHA experiments. 2. Bot interaction datasets generated for adversarial evaluation. The datasets are provided in JSON format and include interaction-level behavioral data used for reinforcement learning–based CAPTCHA adaptation and evaluation. Files: - adaptive-captcha-defense_JSON_Human-Dataset.zip - adaptive-captcha-defense_JSON_Bot-Dataset.zip The datasets support the reproducibility of the experiments and results presented in the associated publication.本仓库包含用于研究《基于强化学习设计自适应防御型验证码（CAPTCHA）系统》的数据集。 本仓库涵盖以下内容： 1. 验证码实验过程中采集的人类交互数据集。 2. 为开展对抗性评估而生成的机器人交互数据集。 本数据集采用JSON格式存储，包含用于基于强化学习的验证码自适应开发与效果评估的交互级行为数据。 提供的数据集文件如下： - adaptive-captcha-defense_JSON_Human-Dataset.zip - adaptive-captcha-defense_JSON_Bot-Dataset.zip 本数据集可支撑相关已发表论文中所述实验与结果的可复现性。