InnerCore

These are the supplementary files for the Data Depth and Core-based Trend Detection paper.

Usage

Until this is packaged, source the file(s) algorithms/innerCore.R or utils.py

Get a weighted directed graph with named vertices and edge weights:

In R:

g <- erdos.renyi.game(200, 2/200, directed = T)
E(g)$weight <- 1:ecount(g)
V(g)$name <- paste("v", 1:vcount(g), sep="")

In Python:

g = nx.erdos_renyi_graph(n=200, seed=1, p=2 / 200, directed=True)

Run innerCore with default parameters

> innerCore(g)
     node
  1:   v2
  2:   v3
  3:   v4
  4:   v5
  5:   v6
 ---     
148: v194
149: v195
150: v197
151: v198
152: v200

Run innerCore with builtin edge property functions

> innerCore(g, featureComputeFun = customNodeFeatures(c("indegree", "triangles")))
    node
 1:   v1
 2:   v2
 3:   v3
 4:   v4
 5:   v7
 --- 
31: v168
32: v169
33: v173
34: v183
35: v200

Get the count of each node consistuting a 3-node motif using the utility function in innerCore.R

> countThreeNodeMotifs(g)
  node motif1 motif2 motif3 motif4 motif5 motif6 motif7 motif8 motif9 motif10 motif11 motif12 motif13 motif14 motif15 motif16
1   v1      0      0      0      3      0      0      0      0      0       0       0       0       0       0       0       0
2   v2      0      0      0     15      1     12      0      0      0       0       0       0       0       0       0       0
3   v3      0      4      0      0      0      0      0      0      0       0       0       0       0       0       0       0
4   v4      0      0      0      1      0      2      0      0      0       0       0       0       0       0       0       0
5   v5      0      0      0      0      3      3      0      0      0       0       0       0       0       0       0       0
 ---
196 v196    3      0      0      0      0      0      0      0      0       0       0       0       0       0       0       0
197 v197    0      0      0      6      0      4      0      0      0       0       0       0       0       0       0       0
198 v198    0      0      0      1      1      4      0      0      0       0       0       0       0       0       0       0
199 v199    0      0      0      1      3      6      0      0      0       0       0       0       0       0       0       0
200 v200    0      0      0      3      3      9      0      0      0       0       0       0       0       0       0       0

Get the centered 3-node motif counts of each node using the utility function in utils.py

>>> centerTriadCensus(g)
{'motif1': {2: 1, 4: 1, 5: 2, 6: 1, 8: 1, 9: 1, 11: 1, 12: 1, 14: 1, 16: 1, 17: 1, 19: 1, 22: 2, 24: 1, 28: 2, 29: 1, 31: 1, 35: 1, 36: 1, 39: 1, 41: 1, 45: 2, 47: 1, 48: 1, 51: 2, 53: 4, 54: 2, 59: 2, 60: 1, 61: 1, 62: 1, 64: 1, 65: 1, 67: 1, 70: 2, 71: 6, 72: 1, 76: 1, 77: 1, 82: 2, 83: 1, 84: 1, 89: 1, 90: 1, 94: 2, 96: 1, 97: 1, 100: 1, 102: 1, 103: 2, 108: 1, 109: 1, 111: 1, 112: 1, 113: 1, 115: 1, 119: 3, 121: 1, 123: 1, 124: 1, 125: 1, 127: 1, 130: 1, 132: 1, 135: 1, 139: 1, 141: 1, 142: 4, 143: 1, 145: 1, 146: 1, 149: 3, 151: 3, 156: 1, 159: 1, 161: 1, 163: 1, 164: 1, 165: 1, 166: 1, 167: 1, 168: 1, 169: 1, 175: 1, 177: 2, 178: 1, 179: 1, 182: 1, 183: 3, 185: 1, 187: 1, 189: 1, 190: 1, 191: 1, 192: 1, 195: 1, 196: 1, 197: 1, 199: 1}, 
'motif6': {}, 
'motif5buy': {74: 1, 10: 1, 187: 1, 31: 1, 52: 1, 158: 2, 175: 3}, 
'motif5sell': {185: 1, 17: 1, 15: 1, 53: 1, 65: 1, 192: 2, 96: 1, 161: 2}, 
'motif4': {0: 1, 1: 1, 6: 1, 7: 1, 15: 15, 19: 1, 21: 1, 23: 2, 26: 1, 27: 1, 30: 4, 31: 9, 32: 1, 33: 1, 34: 1, 38: 1, 39: 3, 40: 1, 44: 1, 46: 1, 49: 1, 50: 1, 52: 5, 54: 1, 55: 1, 56: 1, 57: 1, 58: 1, 59: 3, 63: 1, 67: 1, 69: 1, 70: 1, 71: 3, 73: 1, 74: 1, 76: 3, 79: 1, 80: 1, 81: 1, 85: 6, 87: 1, 92: 1, 93: 1, 95: 1, 98: 1, 99: 2, 102: 1, 104: 1, 105: 1, 106: 3, 107: 1, 110: 3, 114: 1, 116: 1, 117: 1, 118: 1, 120: 1, 126: 3, 133: 1, 134: 2, 135: 3, 136: 1, 137: 1, 138: 1, 140: 1, 141: 9, 143: 1, 144: 1, 147: 1, 148: 1, 149: 1, 151: 1, 152: 1, 153: 1, 154: 1, 155: 1, 157: 1, 159: 1, 160: 1, 166: 3, 171: 2, 172: 1, 173: 3, 174: 6, 175: 10, 176: 1, 180: 1, 183: 1, 184: 5, 186: 1, 187: 1, 185: 1, 188: 3, 189: 1, 194: 1}, 
'motif11': {}}

Computing expansion and decay measures of daily temporal networks

For an example directory with the following files:

eth_innerCore_025_1661126400.csv
eth_innerCore_025_1661212800.csv
eth_innerCore_025_1661299200.csv

where each file contains the addresses of the innercore of a day is in the form:

	node
1	0x91aae0aafd9d2d730111b395c6871f248d7bd728
2	0xa6807d794411d9a80bc435dfc4cda0ba0ddde979
3	0x0084dfd7202e5f5c0c8be83503a492837ca3e95e
4	0x5041ed759dd4afc3a72b8192c143f72f4724081a
5	0xffec0067f5a79cff07527f63d83dd5462ccf8ba4
        ---    

Call computeExpansionDecay from utils.py:

>>> computeExpansionDecay(1, 1661126400, 1661299200, "eth_innerCore_025_", "node")

to produce the results file expansion_decay_1661212800_to_1661299200_i=1.csv in the form:

	timestamp	expansion	decay
0	1661212800	0.387921022	0.516840883
1	1661299200	0.414666667	0.514666667

Compute the motif NF-IAF scores of each address of each day

For an example directory with the following files:

alphaCore_1648857600_motif4.csv
alphaCore_1648944000_motif4.csv
alphaCore_1649030400_motif4.csv

where each file contains the motif counts of each address of each day in the form:

        addresses	                                occurrences
1	0x28c6c06298d514db089934071355e5743bf21d60	847
2	0xdfd5293d8e347dfe59e90efd55b2956a1343963d	799
3	0x56eddb7aa87536c09ccc2793473599fd21a8b17f	577
4	0x876eabf441b2ee5b5b0554fd502a8e0600950cfa	136
5	0xba12222222228d8ba445958a75a0704d566bf2c8	113
        ---    

Call computeNFIAF from utils.py:

>>> computeNFIAF(1648857600, 1649030400, ["motif4"], "alphaCore_", "addresses", "occurrences")

to produce the results file nfiaf_1648857600_to_1649030400_motif4.csv in the form:

	address	                                        timestamp	nfiaf
0	0xba12222222228d8ba445958a75a0704d566bf2c8	1648857600	0.016026962
1	0xba12222222228d8ba445958a75a0704d566bf2c8	1648944000	0.009198891
2	0x5a6a4d54456819380173272a5e8e9b9904bdf41b	1648857600	0.007091576
3	0xba12222222228d8ba445958a75a0704d566bf2c8	1649030400	0.006880386
4	0x7abe0ce388281d2acf297cb089caef3819b13448	1648857600	0.004113114
        --- 

Datasets

The full Ethereum Stablecoin Networks and Ethereum Transaction Network is publicly available on Chartalist.

Citing

Please use the following BibTeX entry:

@ARTICLE{10.3389/fbloc.2024.1342956,
  
AUTHOR={Zhu, Jason and Khan, Arijit and Akcora, Cuneyt Gurcan},   
	 
TITLE={Data depth and core-based trend detection on blockchain transaction networks},      
	
JOURNAL={Frontiers in Blockchain},      
	
VOLUME={7},           
	
YEAR={2024},      
	  
URL={https://www.frontiersin.org/articles/10.3389/fbloc.2024.1342956},       
	
DOI={10.3389/fbloc.2024.1342956},      
	
ISSN={2624-7852},   
   
ABSTRACT={Blockchains are significantly easing trade finance, with billions of dollars worth of assets being transacted daily. However, analyzing these networks remains challenging due to the sheer volume and complexity of the data. We introduce a method named InnerCore that detects market manipulators within blockchain-based networks and offers a sentiment indicator for these networks. This is achieved through data depth-based core decomposition and centered motif discovery, ensuring scalability. InnerCore is a computationally efficient, unsupervised approach suitable for analyzing large temporal graphs. We demonstrate its effectiveness by analyzing and detecting three recent real-world incidents from our datasets: the catastrophic collapse of LunaTerra, the Proof-of-Stake switch of Ethereum, and the temporary peg loss of USDC–while also verifying our results against external ground truth. Our experiments show that InnerCore can match the qualified analysis accurately without human involvement, automating blockchain analysis in a scalable manner, while being more effective and efficient than baselines and state-of-the-art attributed change detection approach in dynamic graphs.}
}