Knowledge Graph for TIme-Series Data

Question

Would storing time series data in a Knowledge Graph be a good idea ? What could be the benefits of doing so ?

Answer 1

It depends on the queries you want to do on the time series data, but I suspect the answer is NO.

Typical queries on time series data include the following:

• moving averages; e.g. 30 day average of stock prices
• median
• accounting functions; e.g. average growth rate, amortization, internal rate of return and so on.
• statistical functions; e.g. autocorrelation, and correlation between two series.
• pattern finding; i.e. find a time series (or multiple time series) that has a similar pattern to this time series

In general time series data have a greater need for aggregation of a collection of data rather creating a graph of the data. This will likely cause any time series related queries to have poor performance on a graph like database.

A factor to consider is that the amount of data stored for time series can be way bigger than that for of a typical knowledge graph depending on the sample rate of the time series data.

Here are some of the references that brought me to this conclusion:

1. Indexing Strategies for Time Series Data
2. Demystifying Graph Databases - Analysis and Taxonomy of Data Organization, System Designs, and Graph Queries