MalDevGenome: A Longitudinal and Large-Scale AndroidMalware Developer Dataset

Dataset for the paper "MalDevGenome: A Longitudinal and Large-Scale AndroidMalware Developer Dataset". We have collected 1,411,830 apps from 10,708 different malware developer. These apps belongs to 1,035 different malware families. We provided the basic information, anti-virus detection result and sensitive behaviors of the apps, including app name, package name, apk file hashes, developer certificate hashes, permission, service, content provider, activity, broadcast receiver, library, hardcode string and the malware family name. And we will upload some representative malware developers’ application APK files later. As far, we uploaded 3 CSV file, including app basic informations, Virustotal's detection results and sensitive behaviors . (1)MalDevGenome_basic_info.csv App basic informations, including app name, package name, developer signature, apk file hashes, AV-Rank, malware family, etc. (2)MalDevGenome_virustotal_detection.csv Detailed VirusTotal's detection, including AV-Rank and details detection information of each anti-virus engine. These data are indexed according to Apk file MD5. (3)MalDevGenome_sensitive_behaviors.csv The sensitive behaviors of the samples, including permission, service, content provider, activity, broadcast receiver, library and hardcode string, etc. These data are indexed according to Apk file MD5.本数据集为论文《MalDevGenome：一款纵向大规模安卓恶意软件开发者数据集》所用数据集。本数据集共收录了来自10708名不同恶意软件开发者的1411830个应用程序，这些应用隶属于1035个不同的恶意软件家族。本数据集提供了应用的基础信息、杀毒引擎检测结果与敏感行为数据，具体包括应用名称、包名、APK文件哈希、开发者证书哈希、权限（Permission）、服务（Service）、内容提供器（Content Provider）、活动（Activity）、广播接收器（Broadcast Receiver）、依赖库、硬编码字符串以及恶意软件家族名称。后续我们将上传部分代表性恶意软件开发者的应用APK文件。截至目前，我们已上传3个CSV格式文件，具体如下：1. MalDevGenome_basic_info.csv：应用基础信息表，包含应用名称、包名、开发者签名、APK文件哈希、AV评级（AV-Rank）、恶意软件家族等字段。2. MalDevGenome_virustotal_detection.csv：VirusTotal详细检测结果表，包含各杀毒引擎的AV评级与详细检测信息，以APK文件MD5作为索引。3. MalDevGenome_sensitive_behaviors.csv：样本敏感行为表，包含权限（Permission）、服务（Service）、内容提供器（Content Provider）、活动（Activity）、广播接收器（Broadcast Receiver）、依赖库以及硬编码字符串等字段，以APK文件MD5作为索引。