Jump to content
Gov  ·  Market  ·  Community  ·  Policies  ·  Funding  ·  Open Calls  ·  Get started

Gov/en/Portal:R&D/Innovations:Rating Balancing Policy: Difference between revisions

From WikiDeal
Remove hard {{NotApproved}} notice (redundant with auto-generated ApprovedRevs notice) — Theo request 2026-06-20
Add links to Maturity/Civic Incentive, and mechanisms for education, early reviewer Karma rewards, and feedback tax
(4 intermediate revisions by the same user not shown)
Line 1: Line 1:
{{KidsIntro|Some people always give 1 star or always give 5 stars. WikiDeal gently balances those extremes so reviews stay fair and honest.}}
{{KidsIntro|Some people always give 1 star or always give 5 stars. WikiDeal gently balances those extremes, and invites many different people to give their opinion, so reviews stay fair and honest.}}
{{ExpertIntro|The Rating Balancing Policy (previously presented as "Judgment Balance Indicator") aims at proactively stimulating balance in marketplace ratings, not just measuring it. The goal is feedback that is honest, calibrated and diversified: coming from many different people, not always the same ones, supported by motivation and incentive tools close to the civic incentive approach. It concerns the marketplaces and pairs with Incentivized Feedback, connecting to the Maturity Score and Civic Incentive by educating users, rewarding early detailed reviews, and taxing systematic ignorance or artificial positivity.}}
__TOC__


== Judgment Balance Indicator ==
= Rating Balancing Policy =
''Innovation WikiDeal R&D''
''Innovation, WikiDeal R&D. Previously presented as "Judgment Balance Indicator": renamed, because this is about rating, not judgment.''


{| class="wikitable"
{| class="wikitable"
|-
|-
| Origin || 🟢 Théo Bondolfi
| Origin || 🟢 WikiDeal concept
|-
|-
| Status || Prototype 1 — In testing
| Status || Under study (July 2026, proposal)
|}
|}


=== What is it? ===
<span id="what-is-it"></span>
The Judgment Balance Indicator is an auto-weighting algorithm that adjusts extreme ratings — such as systematic 1-star or 5-star scores — based on a rater's historical patterns. The goal is to encourage honest, calibrated feedback and to prevent gaming by chronic over-raters or habitual punishers, keeping the evaluation system trustworthy.
== What is it? ==


=== How it works on WikiDeal ===
The Rating Balancing Policy aims at keeping marketplace ratings '''honest, calibrated and diversified'''. It was previously presented as the "Judgment Balance Indicator"; the name changed for two reasons:
Ratings are weighted with reference to each rater's track record, so consistently extreme behaviour carries less distorting weight. This supports fairer aggregate scores across the platform and complements [[Gov/en/Portal:R&D/Innovations:Paid Feedback|Paid Feedback]]. The exact weighting model is documented progressively as the platform evolves.
 
* what is balanced is '''rating''', not judgment;
* it is not just an indicator: the policy aims at '''proactively stimulating balance''', not only measuring it.
 
<span id="honest-calibrated-diversified"></span>
== Honest, calibrated, diversified ==
 
* '''Honest''': ratings should reflect real experiences, without gaming.
* '''Calibrated''': extreme patterns (such as systematic 1-star or 5-star scores) are weighted with reference to each rater's track record, so consistently extreme behaviour carries less distorting weight.
* '''Diversified''': feedback should come from many different people, not always the same ones commenting (and not always the same ones commenting negatively). Motivation and incentive tools, close to the [[Gov/en/Portal:Economy/Civic-Incentive|civic incentive]] approach, aim at widening the circle of people who give feedback.
 
<span id="how-it-works"></span>
== How it works on WikiDeal ==
 
The policy concerns the marketplaces and belongs to the governance family of innovations. It pairs with [[Gov/en/Portal:R&D/Innovations:Incentivized Feedback|Incentivized Feedback]], which stimulates detailed feedback, while the Rating Balancing Policy keeps the resulting ratings fair. It also connects directly with the [[Gov/en/Portal:Economy/Civic-Incentive|Civic Incentive]] and the [[Gov/en/Portal:R&D/Innovations:Maturity Score|Maturity Score]] systems.
 
The exact weighting model and stimulation tools are documented progressively as the platform evolves, structured around three core mechanisms:
 
* '''User Education:''' Users are educated on how to leave constructive, precise comments without exaggerating or wasting time. The system provides guidance on moderation, precision, and contextual use cases to ensure feedback is genuinely useful.
* '''Early Reviewer Advantages:''' For the very first users who take the time to comment on new services, persons, or groups, the platform proactively offers advantages. These early, detailed comments are heavily rewarded—notably in [[Gov/en/Portal:Economy/Karma-Tokens|Karma tokens]] (for example, earning the right to an extra free night in a 7-night accommodation booking). This jumpstarts the reputation and maturity of the given service.
* '''Feedback "Tax" and Penalties:''' To prevent skewed rating distributions and artificial positivity, a "tax" or penalty mechanism is applied to users who too frequently ignore feedback requests, or who exclusively leave positive comments without nuance. This ensures that the overall rating ecosystem remains critical, realistic, and highly reliable.


----
----
'''See also:''' [[Gov/en/Portal:R&D/Innovations:Main|All innovations]] · [[Gov/en/Portal:R&D/Main|R&D Portal]]
'''See also:''' [[Gov/en/Portal:R&D/Innovations:Incentivized Feedback|Incentivized Feedback]] · [[Gov/en/Portal:Economy/Civic-Incentive|Civic incentive]] · [[Gov/en/Portal:R&D/Innovations:Main|All innovations]] · [[Gov/en/Portal:R&D/Main|R&D Portal]]


[[Category:Migration June 2026]]
[[Category:Migration June 2026]]
<!-- visible -->
<!-- visible -->

Revision as of 17:54, 22 July 2026

💡 In simple words: Some people always give 1 star or always give 5 stars. WikiDeal gently balances those extremes, and invites many different people to give their opinion, so reviews stay fair and honest.

🎯 In 20 seconds (scientific summary): The Rating Balancing Policy (previously presented as "Judgment Balance Indicator") aims at proactively stimulating balance in marketplace ratings, not just measuring it. The goal is feedback that is honest, calibrated and diversified: coming from many different people, not always the same ones, supported by motivation and incentive tools close to the civic incentive approach. It concerns the marketplaces and pairs with Incentivized Feedback, connecting to the Maturity Score and Civic Incentive by educating users, rewarding early detailed reviews, and taxing systematic ignorance or artificial positivity.

Rating Balancing Policy

Innovation, WikiDeal R&D. Previously presented as "Judgment Balance Indicator": renamed, because this is about rating, not judgment.

Origin 🟢 WikiDeal concept
Status Under study (July 2026, proposal)

What is it?

The Rating Balancing Policy aims at keeping marketplace ratings honest, calibrated and diversified. It was previously presented as the "Judgment Balance Indicator"; the name changed for two reasons:

  • what is balanced is rating, not judgment;
  • it is not just an indicator: the policy aims at proactively stimulating balance, not only measuring it.

Honest, calibrated, diversified

  • Honest: ratings should reflect real experiences, without gaming.
  • Calibrated: extreme patterns (such as systematic 1-star or 5-star scores) are weighted with reference to each rater's track record, so consistently extreme behaviour carries less distorting weight.
  • Diversified: feedback should come from many different people, not always the same ones commenting (and not always the same ones commenting negatively). Motivation and incentive tools, close to the civic incentive approach, aim at widening the circle of people who give feedback.

How it works on WikiDeal

The policy concerns the marketplaces and belongs to the governance family of innovations. It pairs with Incentivized Feedback, which stimulates detailed feedback, while the Rating Balancing Policy keeps the resulting ratings fair. It also connects directly with the Civic Incentive and the Maturity Score systems.

The exact weighting model and stimulation tools are documented progressively as the platform evolves, structured around three core mechanisms:

  • User Education: Users are educated on how to leave constructive, precise comments without exaggerating or wasting time. The system provides guidance on moderation, precision, and contextual use cases to ensure feedback is genuinely useful.
  • Early Reviewer Advantages: For the very first users who take the time to comment on new services, persons, or groups, the platform proactively offers advantages. These early, detailed comments are heavily rewarded—notably in Karma tokens (for example, earning the right to an extra free night in a 7-night accommodation booking). This jumpstarts the reputation and maturity of the given service.
  • Feedback "Tax" and Penalties: To prevent skewed rating distributions and artificial positivity, a "tax" or penalty mechanism is applied to users who too frequently ignore feedback requests, or who exclusively leave positive comments without nuance. This ensures that the overall rating ecosystem remains critical, realistic, and highly reliable.

See also: Incentivized Feedback · Civic incentive · All innovations · R&D Portal