🧑🦱DAO Reputation System（delete）
Last updated
Last updated
On-chain identity verification is indispensable in Web3, but existing verification methods often fall short in providing a comprehensive reflection of a user's participation in DAOs. Optimizing identity verification is crucial to objectively assess a user's activities and contributions within a DAO.
A DAO reputation system is pivotal for industry development, and the reasons include:
Incentivizing Collaboration and Contribution: By rewarding beneficial contributions and cooperation, members are encouraged to actively participate, fostering a positive community ecosystem.
Enhancing Governance Efficiency: Granting contributors more power and decision-making opportunities increases governance efficiency, enabling organizations to adapt more flexibly to market changes.
Preventing Abuse and Attacks: Introducing a reputation system can mitigate abuse and attacks by establishing a record of member behavior, allowing for punitive measures against those who seek to disrupt the system.
Building Community Trust: A reputation system contributes to building trust within the community, lowering collaboration risks, and improving overall efficiency.
Fostering Community Development: Clear incentive mechanisms attract more participants to the community, driving prosperity and vitality within the community.
DAO BASE is a comprehensive platform dedicated to DAOs, currently the largest DAO repository in the market. The platform aggregates numerous governance frameworks, providing the most comprehensive data on DAOs and meticulously documenting users' various activities within DAOs. Through a unique evaluation system, the platform plans to record users' DAO activities on the blockchain, allocate DAO on-chain reputation scores to each user, and award DAO badges. Reputation scores are primarily based on a comprehensive assessment of users' basic information integrity, proposal creation, proposal voting, delegation, voting power, multisignature, and other aspects. All of this data is stored on the blockchain to ensure transparency and tamper resistance.
DAO badges serve as symbols of users' on-chain reputation data for contributions and behaviors within DAOs. Users can continuously accumulate DAO data by actively participating in DAO activities, thereby increasing their on-chain reputation scores and upgrading badges. This score reflects their status and influence within the DAO community, where higher scores represent more outstanding contributions, greater influence, and an increased likelihood of being selected as a representative or leader.
Users with higher on-chain reputation scores may also benefit from future airdrop activities, with opportunities to acquire more tokens or other privileges. This incentivizes them to actively support the development of DAOs, fostering the healthy growth of DAOs. Therefore, DAO badges and on-chain reputation scores are not only symbols of pride but also mechanisms to reward users for actively participating in and supporting DAOs.
The user's DAO on-chain reputation score is primarily determined based on their contributions in six modules. Each module corresponds to a specific score, and the weights for each module are calculated using the CRITIC algorithm. The final evaluation is derived by combining the scores of each module with their respective weights.
The six major modules considered for the user's DAO on-chain reputation score are as follows:
Basic Information Completeness: First, rank the users based on the completeness of their basic information. Then, calculate the Basic Score using the formula:
Basic Score = (1 - (Rank of Basic Information / Total Number of Users)) * 100
Participation in DAO/Related DAOs: First, rank the users based on the number of DAOs they participate in. Then, calculate the Related DAO Score using the formula:
Related DAO Score = (1 - (Rank of Related DAO / Total Number of Users)) * 100
Proposal Creation: First, rank the users based on the number of proposals they have created. Then, calculate the Created Proposals Score using the formula:
Created Proposals Score = (1 - (Rank of Created Proposals / Total Number of Users)) * 100
Participation in Voting: First, rank the users based on the number of times they have participated in voting. Then, calculate the Votes Score using the formula:
Votes Score = (1 - (Rank of Votes / Total Number of Users)) * 100
Received Delegations: First, rank the users based on the number of addresses currently delegating to them. Then, calculate the Delegations Score using the formula:
Delegations Score = (1 - (Delegations Rank / Total Number of Users)) * 100
Voting weight valuation: First, rank the users based on the product of their voting power percentage in the latest proposal of each DAO they participate in and the market value of the DAO's governance tokens. Then, calculate the Voting Weight Valuation Score using the formula:
Voting Weight Valuation Score = (1 - (Voting Weight Valuation Rank / Total Number of Users)) * 100
Reputation Score = (Voting Weight Valuation Score * Weight) + (Related DAO Score * Weight) + (Created Proposals Score * Weight) + (Votes Score * Weight) + (Delegations Score * Weight) + (Basic Score * Weight)
There are mainly two types of methods for calculating weights commonly seen in the market:
Subjective Methods: Delphi Method, Forced Comparison Method, AHP Hierarchical Analysis Method, Multivariate Analysis Method, Fuzzy Statistical Method, etc.
Objective Methods: Entropy Method, Coefficient of Variation Method, CRITIC Method.
When calculating weights, if there is a deep understanding of different indicators, subjective methods can be chosen for weighting. However, generally, it is necessary to assign weights based on the data within the indicators, using objective methods for calculation.
Objective methods have the following advantages:
Based on data analysis and calculation, the results are more objective and accurate.
Addressing issues with practical data such as outliers, data gaps, etc., can be done purposefully.
The implementation process is standardized, transparent, and easy to repeat and verify.
The Entropy Method requires data to follow a normal distribution and does not consider the correlation between indicators. Since our data does not quite fit a normal distribution, and there might be significant correlations among multiple data dimensions (for instance, the more participation in DAOs, the more proposals created, and possibly more votes cast), the Entropy Method might not be suitable.
The Coefficient of Variation Method requires the importance of indicators to be roughly equivalent. It works better when the importance of each dimension is very close, which might not align with our scenario where subjective differences exist in the importance of different dimensions.
The CRITIC Method, on the other hand, does not have the limitations mentioned above, making it more suitable for our data scenario:
Our data does not necessarily follow a normal distribution.
There are significant correlations among different dimensions of our data.
The entropy method requires data to follow a normal distribution and does not account for the correlation between indicators. On the other hand, the coefficient of variation method requires indicators to have similar importance. In reality, there are indicators that do not follow a normal distribution, and indicators may be correlated. For instance, the number of proposals created in a DAO may increase with the frequency of participation, introducing correlation. Moreover, determining the equal importance of indicators is challenging, which is crucial for the coefficient of variation method. The CRITIC method is employed to overcome these challenges.
The variability of an indicator (Si) is measured by its internal standard deviation. A larger Si indicates that an indicator contains more information, hence requiring a higher weight. The conflict between different indicators (Ri) is calculated by determining the correlation coefficient between them. A higher correlation coefficient suggests stronger interdependence with other indicators, posing redundancy issues. A lower Ri indicates lower correlation with other indicators, allowing for higher weighting. The information content of an indicator (Ci) is represented by Ci = Si * Ri. Weights (Wi) for each indicator are then calculated as Wi = Ci / sum(C), where i = 1,...,n, and Wi represents the weight assigned based on information content.
Example: Indicators X: "A1": [0.1, 0.2, 0.1, 0.35], "A2": [1, 2, 3, 6], "A3": [6, 5, 8, 6], "A4": [100, 100, 10, 20], "A5": [20, 50, 50, 10].
(1) Normalize indicators into X_norm by applying (x_col(i) - min(x_col)) / (max(x_col) - min(x_col)): [[0.0, 0.0, 0.33333, 1.0, 0.25], [0.4, 0.2, 0.0, 1.0, 1.0], [0.0, 0.4, 1.0, 0.0, 1.0], [1.0, 1.0, 0.33333, 0.11111, 0.0]]
(2) Indicator variability S, calculate the standard deviation for each column in X_norm: [0.40926764, 0.37416574, 0.36324158, 0.47385339, 0.44633928]
(3) Indicator conflict R, calculate the correlation coefficient for each column in X_norm and sum 1 - correlation coefficients: [4.39543495, 4.22955306, 4.81258599, 5.71144708, 4.68066049]
(4) Indicator information content C = S * R: [1.79890928, 1.58255385, 1.74813133, 2.70638857, 2.08916262]
(5) Weights W = {Wi = Ci / sum(C), i = 1,...,n}: [0.18124765, 0.15944893, 0.17613155, 0.27267999, 0.21049189]
