Analyzing the propagation mechanism of Web3 narratives with the SIR epidemic model

Using the SIR Epidemic Model to Examine the Propagation of Web3 Narratives

Author: NingNing, Senior Researcher at EMC_Labs, Source: X, @0xNing0x

Today, with the help of Microsoft’s new AI tool Bing, I created something cool: analyzing the propagation mechanism of Web3 narratives based on the epidemiological model SIR.

The SIR model is a classic mathematical model in epidemiology and is one of the most successful and famous models for the spread of infectious diseases.

• Cointelegraph and Chainlink Labs Form the Ultimate Dynamic Duo to Empower Web3 Startups
• Cosmos ecosystem is witnessing a battle for talent, reviewing the events surrounding the competition between Neutron and Osmosis for the Mars Protocol.
• Merit Circle DAO and Immutable: Powering the Playful Revolution!

In the SIR model, the entire population is divided into three groups:

– Susceptible population (S): Individuals who have not yet been infected but lack immunity, making them susceptible to infection after contact with an infected individual.

– Infected population (I): Patients who have already been infected and have the ability to transmit the infection.

– Recovered population (R): Individuals who have recovered from the infection and have gained immunity.

This model not only helps us understand and predict the spread of infectious diseases but also helps us understand and predict the spread of Web3 narratives.

Those who have read “Narrative Economics” understand this point.

Science popularization is over, now let’s start the real performance:

Step 1: Initialization

Susceptible population (S) = Proportion of potential target users for a certain Web3 narrative

Infected population (I) = Proportion of users who already believe in a certain Web3 narrative

Recovered population (R) = Proportion of users who have been desensitized to a certain Web3 narrative

beta = Conversion rate for believing in a certain Web3 narrative

gamma = Conversion rate for desensitizing to a certain Web3 narrative

We set:

S = 0.9, I = 0.1, R = 0.0, beta = 0.8, gamma = 0.01

Step 2: Generate 10000 random numbers, import the SIR model from the Scipy library, and pass our initialization parameters to process the data.**

Step 3: Rearrange the data and visualize the spread of Web3 narratives using a moving bubble chart.**

See the attached image for the visualization result. Under the above initialization conditions, approximately 72% of users will choose to believe in a certain Web3 narrative in the long term, forming a stable “consensus” as commonly mentioned in the cryptocurrency industry.

In addition, I also tested two other sets of initialization conditions:

The first set of Web3 narrative features high transmission rate and high desensitization rate, with initialization conditions: S = 0.9, I = 0.1, R = 0.0, beta = 0.8, gamma = 0.2.

The visualization result shows that only 1% to 3% of users will choose to believe in this set of Web3 narratives in the long term.

The second set of Web3 narrative features medium transmission rate and low desensitization rate, with initialization conditions: S = 0.9, I = 0.1, R = 0.0, beta = 0.5, gamma = 0.01.

Visual results show that 62% to 76% of users will choose to believe in this group of Web3 narratives in the long term.

Conclusion

For a specific Web3 narrative, such as RWA, L2, Web3 games, or NFTs, we can observe and statistically analyze the beta and gamma values in their narrative dissemination to predict whether they can form a long-term stable consensus.

We will continue to update Blocking; if you have any questions or suggestions, please contact us!