pls fix link to data bitcoin_blockchain

[Sandy4321]

pls fix link to data bitcoin_blockchain https://bigquery.cloud.google.com/dataset/bigquery-public-data:bitcoin_blockchain

Anomaly Detection in Blockchain System with TODS

BlockChain: View in Colab

How does TODS detect anomalies from Blockchain System? TODS system can perform three common outlier detection scenarios on time-series data: point-wise detection (time points as outliers), pattern-wise detection (subsequences as outliers), and system-wise detection (sets of time series as outliers).

In this notebook, we use Google BigQuery Bitcoin Blockchain Dataset, which contains information about dates, transactions, blocks and prices of Bitcoin. According to Google, this dataset updates every 10 minutes in the following link: https://bigquery.cloud.google.com/dataset/bigquery-public-data:bitcoin_blockchain

With TODS, we first collect and process the data to put in a dataframe. Next, we use two primitives from TODS to fit and predict on all dimensions of the data, to get the prediction labels and scores. Last, we use the prediction information from TODS primitives to plot and visualize the outliers within the data

[Moirai7]

Are there any specific algorithms/functions to detect subsequence as outliers? Or just label subsequence in the training dataset as outliters and it will automatically fit.