In it’s simplest terms, SimScore API is the middle between a long list of written ideas and enhanced AI synthesis.
In general, AI synthesis is often wrong. It may overrepresent some perspectives, under represent others or simply hallucinate. SimScore provides a consistent statistical baseline that doesn’t shift. This constrains the variability of an AI agent, as an interpreter of consensus patterns rather than a primary synthesizer.
Similarity Score determines the similarity of a pair of ideas.
Similarity Score | Relative Similarity |
---|---|
0.0 | The pair of ideas have no similarity |
>0.2 | The pair of ideas are considered plagiarized |
0.65 - 1.0 | The pair of ideas are essentially duplicates |
Simscore API calculates the Consensus Point from all the written ideas as a mathematical expression.
The similarity of each idea vs the consensus point, expressed as SimScore %.
The API sorts the simscore %’s in order from highest to lowest to determine priority.
Priority# | Idea | Author | SimScore % |
---|---|---|---|
1 | Bias amplification through selective patterns | Emma | 53% |
2 | Contextual misalignment in nuanced reasoning | Liam | 47% |
3 | Temporal flattening of evolving opinions | Sophia | 32% |
4 | Cross-source contamination of unique perspectives | Gabrial | 15% |
In order to see how ideas are related to one another a graph is illustrated and a table is available as an input to AI.
Priority # | X | Y | SimScore% |
---|---|---|---|
1 | 0.515 | -0.287 | 54% |
2 | 0.515 | -0.586 | 51% |
3 | 0.104 | 0.334 | 49% |
4 | -0.580 | 0.044 | 42% |
5 | 0.586 | 0.367 | 33% |
6 | -0.260 | 0.615 | 32% |
7 | -0.506 | -0.406 | 31% |
https://docs.google.com/spreadsheets/d/e/2PACX-1vRbK31Zbther54K6vzp96LYYT5ETFNDalc-uSM_acVyq4GhEz4K1v9jkmg0tuYSM0FucUyvGi8p4Jv3/pubchart?oid=471457277&format=interactive