这里介绍四篇论文提取比特币地址或者实体特征的方法，作为参考。
0x00 Multi-Class Bitcoin-Enabled Service Identification Based on Transaction History Summarization
–2018 IEEE International Conference on Internet of Things (iThings) and IEEE Green Computing and Communications (GreenCom) and IEEE Cyber, Physical and Social Computing (CPSCom) and IEEE Smart Data (SmartData) 两颗星
一天的交易频率：一些服务经常发生交易，一些交易较少
recieved transaction占该地址所有交易的比例
coinbase transaction占该地址所有交易的比例：挖矿类型有更高的比例
交易输入的个数
交易输出的个数
花费多少金额作为输入的频率
花费多少金额作为输出的频率
地址同时出现在输入输出的交易比例
结合了服务类型的特征，考虑地址向量的抽取
0x01 An Evaluation of Bitcoin Address Classification based on Transaction History Summarization
–2019 ICBC IEEE International Conference on Blockchain and Cryptocurrency 三颗星但不是C类
这篇论文基于上一篇论文增加了一些特征
- Basic Statics
- 基于上文的8个特征
- Extra Statics
- lifetime
- total received bitcoins
- total spent bitcoins
- 交易数量，不同类型spent、received、coinbase、payback交易的数量
- balance
- Moments–temporal information corresponding to transaction distributions
- first moment–mean
- second central moment–variance 方差
- third standardized momemt–skewness 偏度
- fourth standardized moment–kurtosis 峰度
有图有真相
0x02 A Probabilistic Model of the Bitcoin Blockchain
–2018 CVPR Workshop paper A类workshop
Address features(10)
- total BTC received
- total BTC balance
- the number of input/output transactions
- address spent/received/payback transaction
Entity features(8) defined at the entity level
Temporal features(16)
- the number of weeks,months,years of activity
- the number of entity traded with per week,month,year
- the number of receiving,sending days
- the activity period duration
- the active day ratio
Centrality features(42) based on entity graph
Motif features
- 1-motif features(44)
- the value of incoming/outgoing transaction
- the number of incoming/outgoing address
- the number of incoming/outgoing transactions per day
- total value
- total fee
- 2-motif features(81)
- 3-motif features(114)
0x03 Identifying Bitcoin Users Using Deep Neural Network
–ICA3PP 2018: Algorithms and Architectures for Parallel Processing C类
address statistical features
- the average time interval of an address be used
- the most frequent transaction type an address involved in
address transaction history features
It directly shows that the patterns of an owner when interacting with the Bitcoin system.
For a simple instance, if an address has never paid any fees other than Bitcoin enforcement in its whole transaction history, it is highly unlikely that it belongs to a user that constantly pay extra transaction fees out of self-willingness to speed up the transaction con rmation.
0x04 Our Features Extracting
基于API提供的JSON数据进一步解析之后，我们可以获取到的数据如下：
Address features
- 交易总数
- 作为输入的交易数，作为输出的交易数，既输入又输出的交易数
- 总收到的BTC
- 总花费的BTC
- 余额
Temporal features
- 一天、一周、一月的交易数
- 交易的平均时间间隔
- 最早交易和最晚交易的时间跨度
- 交易活跃天数的比例
1-motif features
- 交易的输入/输出个数
- 交易费
- 输入/输出金额
Graph Topology features
基于图结构的社区发现